Data transmission method, electronic device, storage medium, computer program product, chip system and data transmission system

Abstract

The embodiments of the present application relate to the technical field of communications. Disclosed are a data transmission method, an electronic device, a storage medium, a computer program product, a chip system and a data transmission system. The method may be applied to a second device side, and comprises: at a first moment, acquiring first information of a first data packet; at a second moment, determining a first transmission mode of the first data packet on the basis of the first information; and at a third moment, acquiring the first data packet. Thus, the second device side can transmit the first data packet on the basis of the first transmission mode. The second moment is not earlier than the first moment. The third moment is later than the second moment. Before obtaining the first data packet, the second device side determines, on the basis of the first information corresponding to the first data packet, the first transmission mode for transmitting the first data packet, wherein the first information is used for indicating information related to the first data packet. Therefore, the first transmission mode determined on the basis of the first information can be suitable for transmitting the first data packet, thereby improving the reliability of air interface transmission.

Description

Data transmission methods, electronic devices, storage media, computer program products, chip systems, and data transmission systems

[0001] This application claims priority to Chinese Patent Application No. 202411930508.1, filed on December 23, 2024, entitled "Data Transmission Method, Electronic Device, Storage Medium, Computer Program Product, Chip System and Data Transmission System", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of communications, and more particularly to a data transmission method, electronic device, storage medium, computer program product, chip system, and data transmission system. Background Technology

[0003] With the continuous development of communication technology, data transmission latency is constantly decreasing and data transmission capacity is constantly increasing. This enables communication technology to be applied to multimedia transmission services that require high real-time performance and large data volumes. For example, multimedia transmission services may include at least one of the following: video transmission, image transmission, audio transmission, or haptic transmission.

[0004] During data transmission, base stations can be used to transmit data packets. However, when data packet sizes fluctuate significantly, it becomes difficult to provide reliable air interface transmission. Summary of the Invention

[0005] This application provides a data transmission method, electronic device, storage medium, computer program product, chip system, and data transmission system to improve the reliability of air interface transmission.

[0006] Firstly, a data transmission method is provided, which can be applied to a second device side. The second device side can be an access side. The access side can include RAN (Radio Access Network) nodes, modules (e.g., circuits, chips, or chip systems) in the RAN nodes, logic nodes capable of implementing all or part of the access side functions, logic modules capable of implementing all or part of the access side functions, or software capable of implementing all or part of the access side functions. The RAN node can include a base station. The base station can be a base station in a 4G, 5G, 5.5G, 6G, or future communication system.

[0007] The method may include: at a first moment, acquiring first information, wherein the first information is used to indicate relevant information of a first data packet; at a second moment, determining a first transmission mode for the first data packet based on the first information; the second moment being no earlier than the first moment; the first transmission mode being used to transmit the first data packet according to the first transmission mode if the first data packet is acquired at a third moment, where the third moment is later than the second moment.

[0008] According to the embodiments of this application, since the second device determines a first transmission method for transmitting the first data packet based on the first information corresponding to the first data packet before acquiring the first data packet, and the first information is used to indicate information related to the first data packet, the first transmission method determined based on the first information can be applied to transmitting the first data packet. In this case, when the first data packet is acquired, the first data packet is transmitted according to the first transmission method previously determined to be applicable to transmitting the first data packet, thereby improving the reliability of air interface transmission.

[0009] In one possible implementation, the first transmission mode can be used to indicate at least one of the following: transmission time, resource size used, transmission order, frequency, time slot, transmit power, channel bandwidth, available resource block (RB), modulation and coding scheme (MCS), or antenna selection scheme.

[0010] In another possible implementation, the first transmission method can also be used to indicate resource allocation information. The resource allocation information may refer to time resources and / or frequency resources allocated to the second device side, etc.

[0011] In one possible implementation, the first transmission method can be used to transmit the first data packet according to the second transmission method when the first data packet is acquired at a third time. The second transmission method can be determined based on the first transmission method and the first data packet.

[0012] Since the first information can be predicted, and predictions may contain errors, while the first data packet acquired at the third moment is the actual data packet, adjusting the first transmission method based on the actual acquired first data packet yields the second transmission method, thus improving its accuracy. In this case, transmitting the first data packet according to the second transmission method further enhances the reliability of air interface transmission.

[0013] In one possible implementation, obtaining the first information may include: determining the first information based on a first neural network model, according to second information and / or third information. The second information may be related to the first information. The third information may be used to indicate at least one of the following: first channel information or first available resources. The first channel information may be used to indicate information related to the first channel. The first channel may be used to transmit a first data packet. The first available resources may be used to indicate available resources on the second device side.

[0014] Since the first information can be obtained by processing the second information using the first neural network model, and the second information is related to the first information, the prediction accuracy of the first neural network model is relatively high. Therefore, the accuracy of the first information is improved, thereby improving the reliability of air interface transmission.

[0015] Since the third information can be used to indicate at least one of the first channel information or the first available resource, the third information embodies information related to the first transmission mode. Based on this, the first information can be obtained by processing the second and third information using the first neural network model. The first neural network model has high prediction accuracy, thus further improving the accuracy of the first information and thereby improving the reliability of air interface transmission.

[0016] In one possible implementation, the first available resource can be used to indicate at least one of the following: a first time-frequency resource, first network congestion information, first spectrum resource, first time resource, first power resource, first antenna resource, first code resource, first computing resource, first storage resource, first network bandwidth, first physical resource block, or first channel resource, etc. The first network congestion information can be used to determine the predicted arrival time corresponding to the first data packet. The first time-frequency resource can be organized into resource blocks. Each resource block has its own size in the time and frequency domains.

[0017] In another possible implementation, obtaining the first information may include: receiving fourth information from the first device side, and determining the first information based on the fourth information. The fourth information may be related to the first information.

[0018] In another possible implementation, obtaining the first information may include receiving the first information from the first device side.

[0019] In one possible implementation, the second information may be determined by the first device based on the fourth information. The fourth information may be used to indicate at least one of the following: a first complexity or a first instruction. The first complexity may be used to indicate the complexity of the first screen. The first instruction may be used to indicate the instruction that triggers the first screen. In addition, the fourth information may also be used to indicate other content that is related to the first information.

[0020] Since the first complexity can be used to indicate the complexity of the first screen, the first instruction can be used to indicate the instruction that triggers the first screen, and the fourth information can be used to indicate at least one of the first complexity or the first instruction, the fourth information is more comprehensive and has a high correlation with the first information. Based on this, the first device determines the second information according to the fourth information, thus improving the accuracy of the second information. Because the first information is determined based on the second information, the accuracy of the first information is further improved.

[0021] In another possible implementation, the first information may be determined by the first device side based on the fourth information.

[0022] Since the first information is determined by the first device based on the fourth information, the amount of data processing on the second device is reduced, thus saving computing resources.

[0023] In one possible implementation, when the first data packet is a video data packet or an image data packet, the first complexity can be used to indicate at least one of the following: a first image entropy, a first image variance, a first edge density, a first gradient magnitude, a first image contrast, a first color contrast, a first color histogram, a first color entropy, or a first fractal dimension. Furthermore, the first complexity can also be used to indicate other parameters related to the generation of the first data packet.

[0024] Since the first complexity can be represented by at least one of the first image entropy, the first image variance, the first edge density, the first gradient magnitude, the first image contrast, the first color contrast, the first color histogram, the first color entropy, or the first fractal dimension, the comprehensiveness of the first complexity is improved, thereby improving the accuracy of the fourth information.

[0025] In the case that the first data packet is an audio data packet, the first complexity can be used to indicate at least one of the following: first audio entropy, first spectral entropy, first dynamic range, or first audio variance.

[0026] Since the first complexity can be represented by at least one of the first audio entropy, the first spectral entropy, the first dynamic range, or the first audio variance, the comprehensiveness of the first complexity is improved, thereby improving the accuracy of the fourth information.

[0027] In one possible implementation, the second information can be used to indicate at least one of the following: the predicted size corresponding to the first data packet, a first prediction confidence level, a first prediction confidence interval, a prediction priority, a predicted arrival time, a second prediction confidence level, a second prediction confidence interval, a predicted service type, or a predicted quality of service requirement. The first prediction confidence level can be used to indicate the confidence level of the predicted size. The first prediction confidence interval can be used to indicate the confidence interval of the predicted size. The second prediction confidence level can be used to indicate the confidence level of the predicted arrival time. The second prediction confidence interval can be used to indicate the confidence interval of the predicted arrival time. The prediction type can be used to indicate the predicted service type of the first data packet. The predicted quality of service requirement can be used to indicate the predicted quality of service required for transmitting the first data packet. And / or, the first information can be used to indicate at least one of the following: the predicted size corresponding to the first data packet, a first prediction confidence level, a first prediction confidence interval, a prediction priority, a predicted arrival time, a second prediction confidence level, a second prediction confidence interval, a prediction type, or a predicted quality of service requirement;

[0028] The first and second pieces of information can be represented in different ways. For example, the first information can be unencrypted, while the second information can be encrypted. The content of the second and first information can also be the same.

[0029] Since the second information can be used to indicate at least one of the following: the predicted size, the first prediction confidence level, the first prediction confidence interval, the prediction priority, the predicted arrival time, the second prediction confidence level, the second prediction confidence interval, the prediction type, or the predicted quality of service requirement corresponding to the first data packet, and the first information can be used to indicate at least one of the following: the predicted size, the first prediction confidence level, the first prediction confidence interval, the prediction priority, the predicted arrival time, the second prediction confidence level, the second prediction confidence interval, the prediction type, or the predicted quality of service requirement corresponding to the first data packet, the second information is more comprehensive. Furthermore, the second information can be the same as the content of the first information, thus having a high correlation with the first information. On this basis, since the first information can be determined based on the second information, the accuracy of the first information is improved.

[0030] Since the second information can be encrypted, the security of information exchange between the second device and the first device is improved.

[0031] In one possible implementation, the first neural network model can be obtained by:

[0032] In one possible implementation, the first neural network model can be a first part of a third neural network model, which is obtained through the following means:

[0033] The third neural network model can be trained on either the second device side or the first device side using the fifth and sixth information. The fifth information can be used to indicate at least one of the following: a second complexity or a second instruction. The second complexity can be used to indicate the complexity of the second screen. The second instruction can be used to indicate the instruction that triggers the second screen. The fifth information can also be used to indicate other content, as long as it is related to the sixth information. The sixth information can be used to indicate actual information related to the second data packet. The sixth information can serve as tag information.

[0034] Since the second complexity can be used to indicate the complexity of the second screen, the second instruction can be used to indicate the instruction that triggers the second screen, and the fifth information can be used to indicate at least one of the second complexity or the second instruction, the fifth information is more comprehensive and has a high correlation with the sixth information. Based on this, the fifth and sixth information are used to train the third neural network model, improving the prediction accuracy of the third neural network model. Since the first neural network model is the first part of the third neural network model, the prediction accuracy of the first neural network model is also improved.

[0035] In another possible implementation, the third neural network model can be trained on either the second device side or the first device side using the fifth, sixth, and seventh information. The seventh information can be used to indicate at least one of the following: second channel information or second available resources. The second channel information can be used to indicate information related to the second channel. The second channel can be used to transmit a second data packet. The second available resources can be used to indicate available resources on the second device side or available resources on other second device sides.

[0036] Since the seventh information can be used to indicate at least one of the second channel information or the second available resource, the seventh information embodies information related to the first transmission mode. Based on this, the third neural network model can be trained using the fifth, seventh and sixth information, thus further improving the prediction accuracy of the third neural network model.

[0037] In another possible implementation, the first neural network model can be trained on either the second device side or the first device side using the eighth and sixth information. The eighth information can be used to indicate information related to the sixth information.

[0038] In another possible implementation, the first neural network model can be trained on the second device side or the first device side using the eighth, seventh, and sixth information.

[0039] In one possible implementation, the second available resource can be used to indicate at least one of the following: a second time-frequency resource, second network congestion information, a second spectrum resource, a second time resource, a second power resource, a second antenna resource, a second code resource, a second computing resource, a second storage resource, a second network bandwidth, a second physical resource block, or a second channel resource, etc. The second network congestion information can be used to determine the predicted arrival time corresponding to the second data packet.

[0040] In one possible implementation, when the second data packet is a video data packet or an image data packet, the second complexity can be used to indicate at least one of the following: second image entropy, second image variance, second edge density, second gradient magnitude, second image contrast, second color contrast, second color histogram, second color entropy, or second fractal dimension. When the second data packet is an audio data packet, the second complexity can be used to indicate at least one of the following: second audio entropy, second spectral entropy, second dynamic range, or second audio variance. And / or

[0041] Since the second complexity can be represented by at least one of the second image entropy, second image variance, second edge density, second gradient magnitude, second image contrast, second color contrast, second color histogram, second color entropy, or second fractal dimension, the comprehensiveness of the second complexity is improved, thereby improving the accuracy of the fourth information.

[0042] Since the second complexity can be represented by at least one of the second audio entropy, the second spectral entropy, the second dynamic range, or the second audio variance, the comprehensiveness of the second complexity is improved, thereby improving the accuracy of the fourth information.

[0043] The sixth piece of information can be used to indicate at least one of the following: the actual size corresponding to the second data packet, a first actual confidence level, a first actual confidence interval, an actual priority, an actual arrival time, a second actual confidence level, a second actual confidence interval, an actual type, or an actual quality of service requirement. The first actual confidence level can be used to indicate the confidence level of the actual size. The first actual confidence interval can be used to indicate the confidence interval of the actual size. The second actual confidence level can be used to indicate the confidence level of the actual arrival time. The second actual confidence interval can be used to indicate the confidence interval of the actual arrival time. The actual type can be used to indicate the service type of the actual first data packet. The actual quality of service requirement can be used to indicate the actual quality of service required for transmitting the second data packet. And / or

[0044] The eighth piece of information may be used to indicate at least one of the following: the actual size corresponding to the second data packet, the first actual confidence level, the first actual confidence interval, the actual priority, the actual arrival time, the second actual confidence level, the second actual confidence interval, the actual type, or the actual quality of service requirement.

[0045] The sixth and eighth messages can be represented in different ways. For example, the sixth message can be unencrypted, while the eighth message can be encrypted. The content of the eighth and sixth messages can also be the same.

[0046] Since the eighth information can be used to indicate at least one of the following corresponding to the second data packet: actual size, first actual confidence level, first actual confidence interval, actual priority, actual arrival time, second actual confidence level, second actual confidence interval, actual type, or actual quality of service requirement, and the sixth information can be used to indicate at least one of the following corresponding to the second data packet: actual size, first actual confidence level, first actual confidence interval, actual priority, actual arrival time, second actual confidence level, second actual confidence interval, actual type, or actual quality of service requirement, the eighth information is more comprehensive. Furthermore, the eighth information can be identical to the sixth information, thus exhibiting a high correlation with the sixth information. Based on this, training the first neural network model using the eighth and sixth information, or training the first neural network model using the eighth, seventh, and sixth information, improves the prediction accuracy of the first neural network model.

[0047] Since the eighth piece of information can be encrypted, the security of information exchange between the second device and the first device is improved.

[0048] In one possible implementation, when the first neural network model is trained by the second device, the second device can send a first signaling message to the first device. The first signaling message can be used to instruct the first device to send fifth and sixth information to the second device via a first method.

[0049] The second device sends a first signaling message to the first device via a first method to obtain the fifth and sixth information, thus achieving the acquisition of the fifth and sixth information.

[0050] In one possible implementation, where the first device side is the serving side and the second device side is the access side, the first approach may include at least one of the following: a GPRS tunneling protocol GTP-U header for the user plane or a first protocol between the access side and the serving side. The first protocol may include fields for indicating fifth information and sixth information. Alternatively...

[0051] When the first device side is the terminal side and the second device side is the access side, the first method may include at least one of the following: a Media Access Control Layer (MAC) CE or a second protocol between the access side and the terminal side. The second protocol includes fields that can be used to indicate fifth information and sixth information.

[0052] By using a custom first or second protocol to transmit the fifth and sixth information, the scalability of the data transmission method in the embodiments of this application is improved.

[0053] In one possible implementation, the first signaling can also be used to indicate the size of the fifth information and the size of the sixth information.

[0054] Since the first signaling can also be used to indicate the size of the fifth and sixth information, it is possible to control the information size, reduce the amount of information transmitted, and save costs.

[0055] When the first device side is the terminal side and the second device side is the access side, the first signaling can also be used to indicate the transmission period. The transmission period can be used to indicate the period during which the terminal side sends the fifth and sixth information to the access side.

[0056] By sending the fifth and sixth messages to the second device according to the transmission cycle, the predictability of information transmission is achieved.

[0057] In one possible implementation, the second information may be received by the second device from the first device via a second method.

[0058] When the first device is the service side and the second device is the access side, the second method may include at least one of the following: a GTP-U header or a third protocol between the access side and the service side. The third protocol may include fields for indicating the second information.

[0059] When the first device side is the terminal side and the second device side is the access side, the second method may include at least one of the following: uplink control information (UCI), user assistance information (UAI), or a fourth protocol between the access side and the terminal side. The fourth protocol may include fields for indicating the second information.

[0060] By using a custom third or fourth protocol to transmit the second information, the scalability of the data transmission method in the embodiments of this application is improved.

[0061] In one possible implementation, when the first neural network model is trained by the second device, the second device can send a second signaling message to the first device. The second signaling message can be used to indicate a second mode and interaction information between the second device and the first device. Alternatively, the second signaling message can be used to indicate the second mode, interaction information, and the magnitude of the interaction information. The second information can be interaction information. The magnitude of the interaction information can be aimed at minimizing transmission.

[0062] It should be noted that the embodiments of this application do not limit the form of the interaction information. For example, the interaction information can be interaction token information. The interaction token information may include at least one interaction token. The size of the interaction token can be optimized to minimize the number of tokens transmitted.

[0063] By defining the meaning, size, and second method of carrying the interaction information between the second and first devices, the security of the interaction between them is improved. Furthermore, since the size of the interaction information can be defined, the information size can be controlled, reducing the amount of information transmitted and saving overhead.

[0064] In one possible implementation, the second device can send a second signaling message to the first device via a third method.

[0065] When the first device is the serving side and the second device is the access side, the third method may include at least one of the following: a GTP-U header or a fifth protocol between the access side and the serving side. The fifth protocol may include fields for indicating the second signaling.

[0066] When the first device side is the terminal side and the second device side is the access side, the third method may include at least one of the following: MAC CE, Downlink Control Information (DCI), or a sixth protocol between the access side and the terminal side. The sixth protocol may include fields for indicating the second signaling.

[0067] By using a custom fifth or sixth protocol to transmit the second signaling, the scalability of the data transmission method in the embodiments of this application is improved.

[0068] By establishing a source graph between the second device and the first device, the second device can transmit data packets according to a pre-determined first transmission method based on an advance awareness of the future size of the first data packet, thereby improving the reliability of air interface transmission and enhancing user experience. In this scenario, the number of users that the second device can serve is increased under limited air interface resources, thus improving air interface resource utilization. The source graph may include at least one of the following: a first neural network model deployed on the second device, a first neural network model deployed on the first device, interaction information between the second and first devices, a second method carrying the interaction information, a fifth piece of information, a sixth piece of information, a seventh piece of information, or an eighth piece of information, etc.

[0069] Secondly, a data transmission method is provided, which can be applied to a first device side. The first device side can be a server side or a terminal side. The server side can include a server, modules in the server (e.g., circuits, chips, or chip systems), logical nodes capable of implementing all or part of the server side functions, logical modules capable of implementing all or part of the server side functions, or software capable of implementing all or part of the server side functions. The terminal side can include a terminal device, modules in the terminal device (e.g., circuits, chips, or chip systems), logical nodes capable of implementing all or part of the terminal side functions, logical modules capable of implementing all or part of the terminal side functions, or software capable of implementing all or part of the terminal side functions. The server or terminal device can be a server or terminal device in a 4G, 5G, 5.5G, 6G, or future communication system. The method can include:

[0070] In one possible implementation, second information is sent to a second device. The second information is used by the second device to acquire first information at a first time, the first information being determined by the second device based on a first neural network model and according to the second information and / or third information. At a second time, a first transmission method for the first data packet is determined based on the first information. The second time is no earlier than the first time. The first transmission method is used to transmit the first data packet if it is acquired at a third time, the third time being later than the second time.

[0071] In one possible implementation, fourth information is sent to the second device side, wherein the fourth information is used by the second device side to obtain first information at a first moment, the first information being determined by the second device side based on the fourth information, and at a second moment to determine a first transmission method for the first data packet based on the first information, the first transmission method being used to transmit the first data packet according to the first transmission method if the first data packet is obtained at a third moment.

[0072] In one possible implementation, first information is sent to a second device, wherein the first information is used by the second device to obtain the first information at a first moment, and at a second moment, to determine a first transmission method for the first data packet based on the first information.

[0073] The first information is used to indicate relevant information of the first data packet, the second information is used to indicate information related to the first information, the third information is used to indicate at least one of the following: first channel information or first available resources, the first channel information is used to indicate information related to the first channel, the first channel is used to transmit the first data packet, the first available resources are used to indicate available resources on the second device side, and the fourth information is related to the first information.

[0074] In one possible implementation, the first transmission mode can be used to indicate at least one of the following: transmission time, resource size used, transmission order, frequency, time slot, transmit power, channel bandwidth, available resource block, modulation and coding scheme, or antenna selection scheme.

[0075] In another possible implementation, the first transmission method can also be used to indicate resource allocation information. The resource allocation information may refer to time resources and / or frequency resources allocated to the second device side, etc.

[0076] In one possible implementation, the first transmission method can be used to transmit the first data packet according to the second transmission method when the first data packet is acquired at a third time. The second transmission method can be determined based on the first transmission method and the first data packet.

[0077] In one possible implementation, the first available resource can be used to indicate at least one of the following: a first time-frequency resource, first network congestion information, a first spectrum resource, a first time resource, a first power resource, a first antenna resource, a first code resource, a first computing resource, a first storage resource, a first network bandwidth, a first physical resource block, or a first channel resource, etc. The first network congestion information can be used to determine the predicted arrival time corresponding to the first data packet.

[0078] In one possible implementation, the second information is determined based on fourth information, which indicates at least one of the following: a first complexity or a first instruction, wherein the first complexity indicates the complexity of the first screen, and the first instruction indicates the instruction that triggers the first screen. Furthermore, the fourth information may also indicate other content that is related to the first information.

[0079] In one possible implementation, the first information is determined based on the fourth information.

[0080] In one possible implementation, when the first data packet is an audio data packet, the first complexity is used to indicate at least one of the following: first image entropy, first image variance, first edge density, first gradient magnitude, first image contrast, first color contrast, first color histogram, first color entropy, or first fractal dimension.

[0081] In the case that the first data packet is an audio data packet, the first complexity is used to indicate at least one of the following: first audio entropy, first spectral entropy, first dynamic range, or first audio variance.

[0082] In one possible implementation, the second information is used to indicate at least one of the following: the prediction size corresponding to the first data packet, the first prediction confidence level, the first prediction confidence interval, the prediction priority, the prediction arrival time, the second prediction confidence level, the second prediction confidence interval, the prediction type, or the prediction quality of service requirement. And / or

[0083] The first information is used to indicate at least one of the following: the prediction size corresponding to the first data packet, the first prediction confidence level, the first prediction confidence interval, the prediction priority, the prediction arrival time, the second prediction confidence level, the second prediction confidence interval, the prediction type, or the prediction quality of service requirement.

[0084] The first and second pieces of information are represented in different ways.

[0085] In one possible implementation, the second information is determined based on the fourth information, and may include: the second information is obtained by inputting the fourth information into a third neural network model, wherein the third neural network model is obtained in the following manner:

[0086] The third neural network model is the second part of the second neural network model, which was obtained in the following way:

[0087] The third neural network model is trained on either the second device side or the first device side using the fifth and sixth information. The fifth information indicates at least one of the following: a second complexity or a second instruction. The second complexity indicates the complexity of the second screen. The fifth information can also indicate other content, as long as it is related to the sixth information. The second instruction indicates the instruction that triggers the second screen, and the sixth information indicates the actual information related to the second data packet.

[0088] The third neural network model is trained on the second device side or the first device side using the fifth, sixth, and seventh information, wherein the seventh information is used to indicate at least one of the following: second channel information or second available resources; the second channel information is used to indicate information related to the second channel; the second channel is used to transmit the second data packet; and the second available resources are used to indicate available resources on the second device side or other available resources on the second device side.

[0089] The second neural network model is trained on either the second device or the first device using the fifth and eighth information. The eighth information is used to indicate information related to the sixth information.

[0090] The second neural network model is trained on the second device side or the first device side using the fifth, seventh, and eighth information.

[0091] In one possible implementation, the second available resource can be used to indicate at least one of the following: a second time-frequency resource, second network congestion information, a second spectrum resource, a second time resource, a second power resource, a second antenna resource, a second code resource, a second computing resource, a second storage resource, a second network bandwidth, a second physical resource block, or a second channel resource, etc. The second network congestion information can be used to determine the predicted arrival time corresponding to the second data packet.

[0092] In one possible implementation, when the second data packet is a video data packet or an image data packet, the second complexity is used to indicate at least one of the following: second image entropy, second image variance, second edge density, second gradient magnitude, second image contrast, second color contrast, second color histogram, second color entropy, or second fractal dimension. When the second data packet is an audio data packet, the second complexity is used to indicate at least one of the following: second audio entropy, second spectral entropy, second dynamic range, or second audio variance. And / or

[0093] The sixth piece of information is used to indicate at least one of the following: the actual size corresponding to the second data packet, the first actual confidence level, the first actual confidence interval, the actual priority, the actual arrival time, the second actual confidence level, the second actual confidence interval, the actual type, or the actual quality of service requirement.

[0094] and / or

[0095] The eighth piece of information is used to indicate at least one of the following: the actual size corresponding to the second data packet, the first actual confidence level, the first actual confidence interval, the actual priority, the actual arrival time, the second actual confidence level, the second actual confidence interval, the actual type, or the actual quality of service requirement.

[0096] The sixth and eighth pieces of information are represented in different ways.

[0097] In one possible implementation, when the second neural network model is trained by the second device side, the first device side receives a first signaling sent by the second device side, wherein the first signaling is used to instruct the first device side to send fifth and sixth information to the second device side in a first manner.

[0098] In one possible implementation, where the first device side is the service side and the second device side is the access side, the first approach includes at least one of the following: a GPRS tunneling protocol GTP-U header for the user plane or a first protocol between the access side and the service side, the first protocol including fields for indicating fifth information and sixth information.

[0099] When the first device side is the terminal side and the second device side is the access side, the first method includes at least one of the following: a Media Intervention Control Layer Control Unit (MAC CE) or a second protocol between the access side and the terminal side, the second protocol including fields for indicating fifth information and sixth information.

[0100] In one possible implementation, the first signaling is also used to indicate the size of the fifth and sixth information. And / or

[0101] When the first device side is the terminal side and the second device side is the access side, the first signaling is also used to indicate the transmission period, which is used to indicate the period during which the terminal side sends the fifth and sixth information to the access side.

[0102] In one possible implementation, sending the second information to the second device side may include: sending the second information to the second device side via a second method.

[0103] When the first device side is the service side and the second device side is the access side, the second method includes at least one of the following: a GTP-U packet header or a third protocol between the access side and the service side, wherein the third protocol includes a field for indicating the second information.

[0104] When the first device side is the terminal side and the second device side is the access side, the second method includes at least one of the following: uplink control information (UCI), user assistance information (UAI), or a fourth protocol between the access side and the terminal side, wherein the fourth protocol includes a field for indicating the second information.

[0105] In one possible implementation, when the second neural network model is trained by the second device side, the first device side receives a second signaling from the second device side. The second signaling is used to indicate the second mode and the interaction information between the second device side and the first device side, or the second signaling is used to indicate the second mode, the interaction information, and the magnitude of the interaction information, wherein the second information is the interaction information.

[0106] In one possible implementation, the first device receives a second signaling sent by the second device via a third method.

[0107] When the first device is the service side and the second device is the access side, the third method includes at least one of the following: a GTP-U header or a fifth protocol between the access side and the service side, wherein the fifth protocol includes fields for indicating the second signaling.

[0108] When the first device side is the terminal side and the second device side is the access side, the third method includes at least one of the following: MAC CE, downlink control information DCI, or a sixth protocol between the access side and the terminal side, wherein the sixth protocol includes a field for indicating the second signaling.

[0109] Thirdly, a first data transmission device is provided, which can be deployed on a second device side. The first data transmission device may include: an acquisition module, configured to acquire first information at a first moment. The first information is used to indicate relevant information of a first data packet. A determination module, configured to determine a first transmission mode of the first data packet based on the first information at a second moment. The second moment is not earlier than the first moment. The first transmission mode is used to transmit the first data packet according to the first transmission mode if the first data packet is acquired at a third moment, where the third moment is later than the second moment.

[0110] Fourthly, a second data transmission device is provided, which can be deployed on the side of a first device. The second data transmission device may include: a sending module for sending second information to the second device. The second information is used by the second device to acquire first information at a first time, the first information being determined by the second device based on a first neural network model, according to the second information and / or third information; and at a second time, to determine a first transmission method for a first data packet based on the first information, the second time being no earlier than the first time. The first transmission method is used to transmit the first data packet according to the first transmission method if the first data packet is acquired at a third time, the third time being later than the second time.

[0111] The sending module is used to send fourth information to the second device. The fourth information is used by the second device to obtain the first information at a first moment (the first information is determined by the second device based on the fourth information), and at a second moment to determine the first transmission method for the first data packet based on the first information. The first transmission method is used to transmit the first data packet if it is obtained at a third moment. Alternatively...

[0112] The sending module is used to send first information to the second device. The first information is used by the second device to obtain the first information at a first moment, and at a second moment to determine the first transmission mode of the first data packet based on the first information.

[0113] The first information is used to indicate relevant information about the first data packet. The second information is used to indicate information related to the first information. The third information is used to indicate at least one of the following: first channel information or first available resources. The first channel information is used to indicate information related to the first channel. The first channel is used to transmit the first data packet. The first available resources are used to indicate available resources on the second device side. The fourth information is used to indicate information related to the first information.

[0114] Fifthly, an electronic device is provided, which may include a processor coupled to a memory for storing a computer program, and the processor for executing the computer program stored in the memory to cause the electronic device to perform the method as described in the first aspect of the present application; or to cause the electronic device to perform the method as described in the second aspect of the present application.

[0115] Optionally, the electronic device further includes a memory, and the processor is coupled to the memory and can be used to execute instructions in the memory to implement the methods in the first aspect to the second aspect and any of the implementable embodiments of the first aspect to the second aspect. Optionally, the electronic device further includes a communication interface, and the processor is coupled to the communication interface. In the embodiments of this application, the communication interface may be a transceiver, a pin, a circuit, a bus, a module or other type of communication interface, and is not limited thereto.

[0116] A sixth aspect provides a processor, comprising: an input circuit, an output circuit, and a processing circuit. The processing circuit is configured to receive signals through the input circuit and transmit signals through the output circuit, causing the processor to execute the methods described in the first to second aspects and any one of the possible implementations of the first to second aspects.

[0117] In specific implementation, the processor can be one or more chips, the input circuit can be input pins, the output circuit can be output pins, and the processing circuit can be transistors, gate circuits, flip-flops, and various logic circuits. The input signal received by the input circuit can be received and input by, for example, but not limited to, a receiver, and the signal output by the output circuit can be output to, for example, but not limited to, a transmitter and transmitted by the transmitter. Furthermore, the input circuit and the output circuit can be the same circuit, which is used as the input circuit and the output circuit at different times. This application does not limit the specific implementation of the processor and various circuits.

[0118] In a seventh aspect, a computer program product is provided, comprising: a computer program (also referred to as code or instructions) that, when run, causes a computer to perform the methods described in the first to second aspects and any one of the implementable embodiments of the first to second aspects.

[0119] Eighthly, a computer-readable storage medium is provided that stores a computer program (also referred to as code or instructions) that, when executed on a computer, causes the computer to perform the methods described in the first to second aspects and any one of the first to second aspects.

[0120] A ninth aspect provides a chip system applied to an electronic device, the chip system including one or more processors, the one or more processors being configured to invoke computer instructions to cause the electronic device to perform the methods of the first to second aspects and any one of the possible implementations of the first to second aspects.

[0121] In a tenth aspect, a data transmission system is provided, including the aforementioned first data transmission device and second data transmission device.

[0122] It should be understood that the beneficial effects of the features corresponding to the first aspect in the second to tenth aspects can be referred to the relevant description of the first aspect above, and will not be repeated here. Attached Figure Description

[0123] Figure 1 is a schematic diagram of the architecture of the communication system of the information transmission method provided in the embodiment of this application;

[0124] Figure 2 is a schematic diagram of the multimedia data transmission process provided in an embodiment of this application;

[0125] Figure 3 is a schematic diagram of the data transmission method provided in the embodiments of this application;

[0126] Figure 4 is a schematic diagram of the principle of the method for generating first information provided in the embodiments of this application;

[0127] Figure 5A is a schematic diagram illustrating the principle of the construction method based on the neural network model provided in the embodiments of this application;

[0128] Figure 5B is a schematic diagram of the model segmentation method provided in the embodiments of this application;

[0129] Figure 6A is a schematic diagram illustrating the principle of a neural network model training method provided in an embodiment of this application;

[0130] Figure 6B is a schematic diagram illustrating the principle of another neural network model training method provided in the embodiments of this application;

[0131] Figure 6C is a schematic diagram illustrating the principle of another neural network model training method provided in the embodiments of this application;

[0132] Figure 7 is a schematic diagram of the execution entity of the training process provided in the embodiment of this application;

[0133] Figure 8 is a schematic diagram of a first protocol or an example of a first protocol provided in an embodiment of this application;

[0134] Figure 9 is a schematic diagram of an example of the fifth or sixth protocol provided in an embodiment of this application;

[0135] Figure 10 is a schematic diagram illustrating the principle of obtaining first information based on a neural network model according to an embodiment of this application.

[0136] Figure 11 is a flowchart of a data transmission method provided in an embodiment of this application;

[0137] Figure 12 is a flowchart of another data transmission method provided in an embodiment of this application;

[0138] Figure 13 is a flowchart of another data transmission method provided in an embodiment of this application;

[0139] Figure 14 is a flowchart of another data transmission method provided in an embodiment of this application;

[0140] Figure 15A is a flowchart of a neural network model training method provided in an embodiment of this application;

[0141] Figure 15B is a flowchart of a method for obtaining first information provided in an embodiment of this application;

[0142] Figure 15C is a flowchart of another method for obtaining first information provided in an embodiment of this application;

[0143] Figure 16A is a flowchart of another neural network model training method provided in an embodiment of this application;

[0144] Figure 16B is a flowchart of another method for obtaining first information provided in an embodiment of this application;

[0145] Figure 16C is a flowchart of another method for obtaining first information provided in an embodiment of this application;

[0146] Figure 17A is a flowchart of another neural network model training method provided in an embodiment of this application;

[0147] Figure 17B is a flowchart of another method for obtaining first information provided in an embodiment of this application;

[0148] Figure 17C is a flowchart of another method for obtaining first information provided in an embodiment of this application;

[0149] Figure 18A is a flowchart of another neural network model training method provided in an embodiment of this application;

[0150] Figure 18B is a flowchart of another method for obtaining first information provided in an embodiment of this application;

[0151] Figure 18C is a flowchart of another method for obtaining first information provided in an embodiment of this application;

[0152] Figure 19 is a flowchart of another neural network model training method provided in an embodiment of this application;

[0153] Figure 20 is a flowchart of another neural network model training method provided in an embodiment of this application;

[0154] Figure 21 is a flowchart of another neural network model training method provided in an embodiment of this application;

[0155] Figure 22 is a flowchart of another neural network model training method provided in an embodiment of this application;

[0156] Figure 23 is a flowchart of another neural network model training method provided in an embodiment of this application;

[0157] Figure 24 is a flowchart of another neural network model training method provided in an embodiment of this application;

[0158] Figure 25 is a schematic block diagram of a first data transmission device provided in an embodiment of this application;

[0159] Figure 26 is a schematic block diagram of a second data transmission device provided in an embodiment of this application;

[0160] Figure 27 is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0161] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.

[0162] To facilitate understanding of the embodiments of this application, the following points will be explained first.

[0163] I. In the embodiments of this application, "instruction" may include direct instruction, indirect instruction, explicit instruction, or implicit instruction. When describing a certain instruction information for indicating A, it can be understood that the instruction information carries A, directly indicates A, or indirectly indicates A.

[0164] II. In the embodiments of this application, " / " can indicate that the objects before and after it are in an "or" relationship. For example, "A / B" can mean A or B. "And / or" can be used to describe three relationships between the related objects. For example, A and / or B can represent three cases: A alone, A and B simultaneously, and B alone. Here, A and B can be singular or plural.

[0165] III. In the embodiments of this application, "at least one" can refer to one or more. "More than one" can refer to two or more, for example, three, four or more. Similar expressions (e.g., at least one, at least one, etc.) are similar. "At least one of the following," "one or more of the following," or similar expressions can refer to any combination of these items, and can include only a single item or a combination of multiple items. For example, at least one of a, b, or c can represent a, b, or c; a and b; a and c; b and c; a, b, and c. Wherein, a, b, and c can be singular or plural.

[0166] IV. In the embodiments of this application, the various numerical designations are merely for descriptive convenience and are not intended to limit the scope of protection of the embodiments of this application. The magnitude of the sequence numbers involved in the embodiments of this application does not imply the order of execution; the execution order of each process should be determined by its function and internal logic. For example, the terms "first," "second," "third," "fourth," and other various terminology (if present) in the specification, claims, and drawings of the embodiments of this application can be used to distinguish similar objects, rather than necessarily to describe a specific order or sequence. Wherein, such terms can be interchanged where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein, and "first," "second," "third," "fourth," etc., are not necessarily different.

[0167] V. In the embodiments of this application, the words "exemplary," "example," or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary," "example," or "for example" should not be construed as being more preferred or advantageous than other embodiments or design solutions. Specifically, the use of "exemplary," "example," or "for example" is intended to present the relevant concepts in a specific manner to facilitate understanding.

[0168] VI. In the embodiments of this application, "sending information / data" only indicates the direction of information / data transmission, including direct transmission via the device's communication interface (e.g., air interface, etc.). "Sending" can also be understood as the "output" of the module interface. "Sending" can include indirect transmission by the processing unit through the communication interface, that is, after the processing unit outputs information / data through the module interface, it is transmitted to the device's communication interface and then sent out. "Receiving information / data" only indicates the direction of information / data transmission, including direct reception via the communication interface. "Receiving" can also be understood as the "input" of the module interface. "Receiving information / data" can include indirect reception by the processing unit through the communication interface, that is, after the communication interface receives information / data, it is transmitted to the module interface of the processing unit and then input to the processing unit by the module interface. "Sending information / data to... (e.g., terminal device)" can be understood as the destination of the information being the terminal device. It can include sending information / data directly or indirectly to the terminal device. "Receiving information / data from... (e.g., terminal device)" can be understood as the source of the information being the terminal device, and can include receiving information / data directly or indirectly from the terminal device. Information / data may undergo necessary processing, such as format changes, between the source and destination, but the destination can understand the valid information / data from the source. Similar statements in the embodiments of this application can be understood in a similar way, and will not be repeated here.

[0169] VII. In the embodiments of this application, "pre-configuration" may include pre-defined features, such as protocol definitions. "Pre-defined features" can be implemented by pre-storing corresponding codes, tables, or other means that can be used to indicate relevant information in the device (e.g., including various network elements). The embodiments of this application do not limit the specific implementation method.

[0170] 8. In the embodiments of this application, "storage" or "preservation" may refer to storage in one or more memories. The one or more memories may be separately configured or integrated into an encoder or decoder, processor, or communication device. Alternatively, some of the one or more memories may be separately configured, while others may be integrated into a decoder, processor, or communication device. The type of memory can be any form of storage medium, and the embodiments of this application do not limit this.

[0171] 9. In the embodiments of this application, the “protocol” may refer to standard protocols in the field of communication, such as fourth-generation (4G) network protocols, fifth-generation (5G) network protocols, new radio (NR) protocols, 5.5G network protocols, sixth-generation (6G) network protocols, and related protocols applied to future communication systems. The embodiments of this application do not limit this.

[0172] 10. In the embodiments of this application, the terms "comprising", "having", and any variations thereof are intended to cover non-exclusive protection. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to which steps or units are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, systems, or devices.

[0173] XI. The arrows or boxes shown by dashed lines in the schematic diagrams of the accompanying drawings of the embodiments of this application may represent optional steps or optional modules.

[0174] 12. Unless otherwise specified or in case of logical conflict, the terms and / or descriptions of different embodiments of this application are consistent and can be referenced in each other. Technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0175] To facilitate understanding of the technical solutions in the embodiments of this application, some terms involved in the embodiments of this application will be explained below.

[0176] 1. Multimedia transmission services can refer to services that transmit multimedia data over a network. Multimedia data can include at least one of the following: video data, image data, audio data, text data, tactile data, or interactive media data, etc. Interactive media can refer to digital media forms in which users interact with each other in some way. Interaction can include at least one of the following: clicking, touching, swiping, dragging, pinching, unfolding, mouse operation, keyboard operation, pressing, gestures, postures, gaze, haptic feedback, or voice, etc. Tactile data can refer to data related to tactile perception. Tactile data can include at least one of the following: vibratory tactile data, kinematic tactile data, or electrotactile data, etc. Vibratory tactile data can be used to simulate vibrations of a predetermined frequency and / or intensity through motor vibration. For example, in shooting games, vibration simulates the special effects produced when using shooting tools. Kinematic tactile data can be used to simulate the weight or pressure of an object through a kinematic tactile system. Kinematic tactile data may include at least one of the following: velocity or acceleration. Electrotactile data may be used to simulate at least one of the following: temperature change, pressure change, or humidity change, via electrical impulses.

[0177] It should be noted that various application scenarios involve multimedia transmission services. For example, these application scenarios may include at least one of the following: Enhanced Mobile Broadband (eMBB), Ultra-Reliable Low Latency Communications (URLLC), Massive Machine Type Communication (mMTC), eMBB+, URLLC+, or mMTC+. For instance, eMBB may include extended reality (XR). Extended reality may include at least one of the following: Virtual Reality (VR), Augmented Reality (AR), or Mixed Reality (MR). eMBB+ may include at least one of the following: Immersive Cloud XR, Cloud Gaming (CG), haptic and multi-sensory communication, or glasses-free 3D holographic display.

[0178] 2. The communication system may include a server, a core network (CN), a radio access network (RAN), and terminal equipment (TE). The radio access network may include at least one RAN node. The RAN node may be a base station (BS). The server can send data packets to the terminal equipment through the core network and the radio access network. In this embodiment, the data packets may include multimedia data packets. That is, the data packets may include at least one of the following: video data packets, image data packets, audio data packets, text data packets, haptic data packets, or interactive media data packets, etc.

[0179] It should be noted that the communication system described in the embodiments of this application can be a 4G communication system, a 5G communication system, a 5.5G communication system, a 6G communication system, or a future communication system.

[0180] 3. A communication protocol refers to the agreement that two parties must follow to complete communication or provide services; hereinafter referred to as a protocol. A protocol can specify at least one of the following: data format, transmission order, or error handling. A protocol stack can refer to the sum of protocols at each layer in a network. For example, a protocol stack can be divided into the Physical Layer (PHY) L1, the Data Link Layer (DLL) L2, and the Network Layer (NL) L3.

[0181] 4. Signaling can refer to the control information and processes used in a communication system to establish, manage, and terminate communication sessions.

[0182] 5. A source can refer to the sender that generates information. A destination can refer to the receiver that receives information. A channel can refer to the medium or path used to transmit information. Information output from a source can be transmitted to a destination through a channel. For example, a source can be a server, and a destination can be a base station or a terminal device. Optionally, a source can be a terminal device, and a destination can be a base station or a server.

[0183] 6. Bit rate (or bit rate) refers to the amount of data transmitted or processed per unit of time, and can be measured in bits per second (BPS). Bit rate is used to measure the transmission rate of data streams and is widely used in video, audio, network communication, and other fields. Bit rate can include variable bit rate (VBR) or constant bit rate (CBR).

[0184] 7. Channel capacity refers to the maximum transmission rate at which information can be transmitted without errors under predetermined channel conditions.

[0185] 8. Quality of Service (QoS) refers to the performance level that a network can provide to applications, users, or data flows. Parameters affecting QoS can include at least one of the following: packet loss information, latency information, bandwidth information, jitter information, throughput, reliability, or availability. Packet loss information can include at least one of the following: packet loss rate or cumulative packet loss. Jitter information can refer to variations in packet arrival time.

[0186] 9. Source control mechanism refers to a mechanism in which the information source adjusts the bitrate according to the quality of service of service data (e.g., multimedia data) to match changes in network status. The information source can be an application layer information source.

[0187] 10. Air Interface (AI), also known as the air interface, refers to the interface between the wireless access network and the terminal equipment.

[0188] 11. Air interface resources refer to the resources required for data transmission between the base station and the terminal equipment. Air interface resources may include at least one of the following: frequency resources, time resources, code resources, or space resources, etc.

[0189] 12. Channel State Information (CSI) can be used to describe the attributes of a channel. CSI can include at least one of the following: a channel matrix, a Channel Quality Indicator (CQI), a Rank Indication (RI), a Precoding Type Indicator (PTI), or a Precoding Matrix Indicator (PMI). The channel matrix can be used to indicate the mapping between the signal transmitted by the transmitter and the signal received by the receiver. The Channel Quality Indicator can be used to quantify the channel quality and provide the signal-to-noise radio (SNR) or signal-to-interference-plus-noise radio (SINR) experienced by the receiver. The Rank Indicator can be used to indicate the number of independent spatial data streams the channel can support. The Precoding Matrix Indicator can be used to indicate the optimal precoding matrix.

[0190] 13. GTP-U (GPRS Tunneling Protocol User Plane) can be a protocol used in a communication system to transmit user data. The GTP-U header can be a component of the GTP-U protocol data packet, and it includes control information required during transmission.

[0191] 14. MAC CE (Media Access Control Element) can be a method for exchanging MAC layer control information between terminal devices and networks.

[0192] 15. DCI (Downlink Control Information) can be transmitted by the base station on the Physical Downlink Control Channel (PDCCH) and is used to instruct terminal equipment on how to receive and send data.

[0193] 16. UCI (Uplink Control Information) is the control information that a terminal device sends to a base station in the uplink.

[0194] 17. UAI (User Auxiliary Information) can refer to additional information used in a communication system to enhance or support the communication process. This information can be used for a variety of purposes, and may include at least one of the following: optimizing network performance, improving user experience, supporting network management or security functions, etc.

[0195] For ease of understanding, the communication system shown in Figure 1 is used as an example to describe the communication system that can be applied to the embodiments of this application.

[0196] Figure 1 is a schematic diagram of the communication system architecture of the information transmission method provided in this application embodiment. Figure 1 shows a schematic diagram of a possible, non-limiting system architecture.

[0197] As shown in Figure 1, the communication system 100 may include a server 110, a core network 120, a radio access network (RAN) 130, and at least one terminal device 140 (e.g., terminal devices 140_1, 140_2, 140_3, 140_4, 140_5, 140_6, and 140_7, which can be collectively referred to as terminal devices 140). The RAN 130 can be wirelessly connected to the terminal devices 140. The interface between the RAN 130 and the terminal devices 140 can be called an air interface. The RAN 130 can be connected to the core network 120 via wired or wireless means.

[0198] Server 110 can be a server that provides various services. For example, server 110 can be used to encode and / or render multimedia data. Server 110 can be a cloud server, also known as a cloud computing server or cloud host. Server 110 can also be a server for a distributed system or a server that incorporates blockchain technology.

[0199] Core network 120 can refer to core network equipment that provides service support for terminal equipment 140. For example, core network 120 may include at least one of the following: User Plane Function (UPF) entity, Session Management Function (SMF) entity, Access and Mobility Management Function (AMF) entity, Policy Control Function (PCF) entity, Application Function (AF) entity, or Network Exposure Function (NEF) entity. The UPF entity can be used to connect to the data network (DN) and perform functions such as user plane data forwarding, session / flow-level billing statistics, or bandwidth limiting. The SMF entity can be used for session management functions of terminal equipment 140, such as IP (Internet Protocol) address allocation, UPF entity selection, and billing and QoS policy control, for example, user session establishment. The PCF entity can be used for policy management functions such as billing policies and QoS policies. The AMF entity can be used for access management and mobility management of terminal equipment 140. Furthermore, the AMF entity can also be used to transmit user policies between terminal equipment 140 and the PCF entity. The AF entity can be used to transmit application-side requirements to the network side. The NEF entity can be used to provide the AF entity with 3GPP network functions and capabilities, and can also enable the AF entity to provide information to 3GPP network functions. It should be noted that entities in this application can also be referred to as network elements or functional entities. For example, the UPF entity can also be referred to as a UPF network element or a UPF functional entity.

[0200] RAN130 can be a cellular system related to the 3rd Generation Partnership Project (3GPP), such as 4G, 5G, 5.5G, 6G, or future communication systems. RAN130 can also be an Open RAN (O-RAN or ORAN), Cloud Radio Access Network (CRAN), or Wireless Fidelity (WiFi) system. RAN130 can also be a communication system that integrates two or more of the above systems.

[0201] RAN130 can include N RAN nodes, namely RAN node 130_1, ..., RAN node 130_n, ..., RAN node 130_N (collectively referred to as RAN node 130). N can be an integer greater than or equal to 1. n can be an integer greater than or equal to 1 and less than or equal to N. RAN nodes can also be called access network devices, RAN entities, or access nodes, etc., and are used to help terminal device 140 achieve wireless access. In addition, RAN node 130 can also be a wireless relay device or a wireless backhaul device (not shown in Figure 1). Multiple RAN nodes 130 can be nodes of the same type or nodes of different types. In some scenarios, the roles of RAN node 130 and terminal device 140 are equivalent. For example, terminal device 140_6 can be a drone or helicopter, which can be configured as a mobile base station. For terminal device 140 accessing RAN130 through terminal device 140_6, terminal device 140_6 is a base station. However, for RAN node 130_N, terminal device 140_6 is a terminal device. RAN node 130 and terminal device 140 are both referred to as communication devices in some cases. For example, RAN node 130_1 and RAN node 130_n in Figure 1 can be understood as communication devices with base station functions. Terminal device 140_1 and terminal device 140_2 in Figure 1 can be understood as communication devices with terminal functions.

[0202] In one possible scenario, RAN node 130 can be a base station (BS), an evolved NodeB (eNB), an access point (AP), a transmission reception point (TRP), a next-generation NodeB (gNB), a next-generation base station in a 6G mobile communication system, a base station in a future communication system, a wireless relay node, a wireless backhaul node, or an access node in a WiFi system, etc. The base station can be a macro base station, micro base station, pico base station, small cell, relay station, donor node, balloon station, or a radio controller in a CRAN scenario. Optionally, RAN node 130 can also be a server, wearable device, vehicle, or in-vehicle equipment, etc. For example, the access network equipment in a Vehicle to Everything (V2X) connection can be a roadside unit (RSU). All or part of the functions of RAN node 130 in this embodiment can also be implemented through software functions running on hardware, or through virtualization functions instantiated on a platform (e.g., a cloud platform). The RAN node 130 may also be equipped with a communication module, circuit, or chip that performs corresponding communication functions. The RAN node 130 may also be configured with program instructions for performing corresponding communication functions and corresponding program instructions. In this embodiment, the RAN node 130 may also be a logic node, logic module, or software capable of implementing all or part of the functions of the RAN node 130.

[0203] In another possible scenario, multiple RAN nodes 130 collaborate to assist terminal device 140 in achieving wireless access, with different RAN nodes 130 each implementing some of the base station's functions. RAN nodes 130 can be Central Units (CUs), Distributed Units (DUs), CU-Control Plane (CP), CU-User Plane (UP), or Radio Units (RUs), etc. CUs and DUs can be set up separately or included in the same network element, such as a Base Band Unit (BBU). The CU and DU can separate the gNB's protocol layers, with some protocol layer functions centrally controlled by the CU, and the remaining partial or complete protocol layer functions distributed in the DU, which is centrally controlled by the CU. As one implementation, the CU deploys the Radio Resource Control (RRC) layer, Packet Data Convergence Protocol (PDCP) layer, and Service Data Adaptation Protocol (SDAP) layer from the protocol stack. The DU deploys the Radio Link Control (RLC) layer, Media Access Control (MAC) layer, and Physical Layer (PHY) in the protocol stack. Therefore, the CU has RRC, PDCP, and SDAP processing capabilities. The DU has RLC, MAC, and PHY processing capabilities. It is understood that the above functional division is merely an example and does not constitute a limitation on the CU and DU. The RU can be included in radio frequency equipment or radio frequency units, for example, in a Remote Radio Unit (RRU), Active Antenna Unit (AAU), or Remote Radio Head (RRH).

[0204] In different systems, CU (or CU-CP and CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an ORAN system, CU can also be called O-CU (i.e., open CU), DU can also be called O-DU, CU-CP can also be called O-CU-CP, CU-UP can also be called O-CU-UP, and RU can also be called O-RU. For ease of description, the embodiments of this application use CU, CU-CP, CU-UP, DU, and RU as examples. Any of the units among CU (or CU-CP, CU-UP), DU, and RU in the embodiments of this application can be implemented through software modules, hardware modules, or a combination of software modules and hardware modules.

[0205] The core network equipment in core network 120 and the RAN node in RAN 130 can be different physical devices, or they can be the same physical device that integrates core network logical functions and radio access network logical functions.

[0206] Terminal equipment 140 can be a device or module with corresponding communication functions. Terminal equipment can also be referred to as user equipment (UE), access equipment, user unit (SU), user station, mobile station, mobile station (MS), remote station, mobile terminal (MT), remote terminal, mobile device, user terminal, terminal, wireless communication equipment, user agent, or user device, etc.

[0207] Terminal device 140 can be a mobile phone, tablet computer, computer with wireless transceiver capabilities, mobile internet device (MID), virtual reality terminal device, augmented reality terminal device, personal digital assistant (PDA), customer premises equipment (CPE), communication equipment mounted on high-altitude aircraft, drone, helicopter, airplane, ship, robot, robotic arm, terminal device in device-to-device (D2D) communication, terminal device in vehicle external connections, terminal device in industrial control, terminal device in machine-type communication (MTC), terminal device in the Internet of Things (IoT), terminal device in autonomous driving, tactile terminal device, in-vehicle terminal device, terminal device in remote medical care, terminal device in smart grid, terminal device in transportation safety, terminal device in intelligent transportation, or smart city. Terminal devices in smart homes, smart offices, wearable devices, transportation vehicles with wireless communication capabilities, communication modules, or terminal devices in communication systems evolved after 5G are not limited to this category in this application. Wearable devices, also known as wearable smart devices, are a general term for devices that are intelligently designed and developed using wearable technology. For example, wearable devices may include at least one of the following: head-mounted displays (HMDs), glasses, gloves, watches, clothing, or shoes. Wearable devices are portable devices that are worn directly on the body or integrated into the user's clothing or accessories. Wearable devices are not just hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly defined wearable smart devices include those that are feature-rich, large in size, and can achieve complete or partial functions without relying on a smartphone, such as smartwatches or smart glasses, as well as those that focus on a specific type of application function and require the use of other devices such as smartphones, such as various smart bracelets or smart jewelry for vital sign monitoring. For example, smart glasses may include at least one of the following: VR glasses or AR glasses, etc.

[0208] It should be noted that, in this embodiment, the device used to implement the function of server 110 can be a server or a device capable of supporting server 110 in implementing this function, such as a chip system, chip, hardware circuit, software module, or hardware circuit plus software module. This device can be installed on server 110 or used in conjunction with server 110. Similarly, the device used to implement the function of RAN node 130 can be RAN node 130 or a device capable of supporting RAN node 130 in implementing this function, such as a chip system, chip, hardware circuit, software module, or hardware circuit plus software module. This device can be installed on RAN node 130 or used in conjunction with RAN node 130. The device for implementing the function of terminal device 140 can be a terminal device or a device capable of supporting terminal device 140 in implementing this function, such as a chip system, chip, hardware circuit, software module, or hardware circuit plus software module. This device can be installed on terminal device 140 or used in conjunction with terminal device 140. This application does not limit the device form of server 110, RAN node 130, and terminal device 140.

[0209] It should also be noted that the server 110, RAN node 130, and terminal device 140 in this application embodiment can be devices with data transmission capabilities in 4G communication systems, 5G communication systems, 5.5G communication systems, 6G communication systems, or future communication systems. The shape or type of the server 110, RAN node 130, and terminal device 140 in 6G and future communication systems is not limited. Furthermore, this application embodiment does not limit the application scenario of the communication system.

[0210] It should be noted that, in the embodiments of this application, the first device side can be a service side or a terminal side, and the second device side can be an access side. The service side may include a server, modules within the server (e.g., circuits, chips, or chip systems), logical nodes capable of implementing all or part of the service side functions, logical modules capable of implementing all or part of the service side functions, or software capable of implementing all or part of the service side functions. The terminal side may include a terminal device, modules within the terminal device (e.g., circuits, chips, or chip systems), logical nodes capable of implementing all or part of the terminal side functions, logical modules capable of implementing all or part of the terminal side functions, or software capable of implementing all or part of the terminal side functions. The access side may include a RAN node, modules within the RAN node (e.g., circuits, chips, or chip systems), logical nodes capable of implementing all or part of the access side functions, logical modules capable of implementing all or part of the access side functions, or software capable of implementing all or part of the access side functions. The RAN node may include a base station.

[0211] The inventive concept of the embodiments of this application will now be described with reference to the accompanying drawings.

[0212] With the continuous improvement of communication transmission rates, real-time multimedia transmission services have gradually become one of the core services in networks. Multimedia transmission services have high requirements for both transmission latency and transmission rate. For example, for XR transmission services, the packet delay budget (PDB) between the fifth and sixth time points can be 10ms. The fifth time point can refer to the moment when the RAN node sends the data packet. The sixth time point can refer to the moment when the terminal side (e.g., the terminal device) receives the data packet. XR transmission services can refer to the multimedia transmission services involved in XR.

[0213] To facilitate understanding of multimedia transmission services, the multimedia data transmission process will be explained below with reference to Figure 2.

[0214] Figure 2 is a schematic diagram of the multimedia data transmission process provided in an embodiment of this application. It should be noted that the multimedia data transmission process in Figure 2 can be applied to the communication system in Figure 1. As one implementation, the server in Figure 2 can be server 110 in Figure 1. The base station in Figure 2 can be RAN node 130_1 in Figure 1. The terminal device in Figure 2 can be terminal device 140_2 in Figure 1.

[0215] As shown in Figure 2, the terminal device can send an uplink request to the server, which can be used to request multimedia data. The server can respond to the uplink request, obtain the multimedia data, and determine the data packets based on the multimedia data. For example, the server can encode the multimedia data. Encoding can include variable bit rate-based encoding or constant bit rate-based encoding. The server can send data packets to the terminal device through a base station. Upon receiving the data packets, the base station can transmit the data packets to the terminal device via air interface transmission.

[0216] As one implementation method, during multimedia data transmission, the information source (e.g., the server in Figure 2) can use a variable bit rate-based encoding method for encoding. The base station can transmit data packets using a first transmission mode of "best-effort transmission over the air interface," meaning that the base station makes every effort to transmit the data packet upon arrival, and chooses to drop the packet if transmission is difficult to complete. In this case, because the information source uses variable bit rate-based encoding, the data packet size fluctuates significantly. If air interface resources are limited and multiple users need to be served simultaneously, large data packets may lead to the base station being unable to provide reliable air interface transmission, resulting in transmission failure.

[0217] To achieve reliable air interface transmission at base stations, it was found that during multimedia data transmission, the application layer source adjusts the bit rate based on a source control mechanism to match changes in network status. Channel capacity can be used to indicate network status. For example, a good network status results in a high channel capacity, while a poor network status results in a low channel capacity. The source control mechanism refers to the mechanism by which the source adjusts the bit rate based on the quality of service (QoS) of the multimedia data to match changes in network status. This source control mechanism is a "probe-then-adjust" method. For example, if the network status is good, i.e., the channel capacity is high, the source perceives good QoS for the multimedia data, meaning there is no packet loss and latency is stable within a predetermined range. In this case, the source does not need to adjust the bit rate. When channel capacity decreases, the source perceives a decline in QoS, which may include increased packet loss and / or increased latency. In this case, the source needs to adjust the bit rate. However, QoS statistics are periodic, resulting in a response time between the source perceiving a decline in QoS and a decrease in channel capacity. During this response time, because the source does not reduce the bit rate, the quality of service is difficult to improve, resulting in a high packet loss rate and / or high latency. This may lead to the base station receiving large data packets. Given limited air interface resources and the need to serve multiple users simultaneously, the base station, relying on its "best-effort" transmission mode, will experience packet loss, making reliable air interface transmission difficult. Furthermore, this may cause stuttering in multimedia data received by users, further degrading the user experience.

[0218] Furthermore, regarding video transmission in multimedia transmission services, it was found that the reason why reliable air interface transmission is difficult to provide is that: since the size of the data packet is related to the complexity of the picture, the data packet size of different video frames may be different. Therefore, when the base station resources are limited, the base station will find it difficult to provide reliable air interface transmission based on the first transmission mode of "best effort transmission over the air interface".

[0219] Based on the above, it was found that the response lag of the source control mechanism is one of the reasons why base stations have difficulty providing reliable air interface transmission. Therefore, embodiments of this application propose to improve the reliability of air interface transmission by shortening the response time. For example, a first transmission method suitable for transmitting the first data packet can be determined before the base station obtains the first data packet to be transmitted, so that when the first data packet is obtained, the first data packet can be transmitted according to the previously determined first transmission method.

[0220] Next, it is necessary to consider how the base station determines the first transmission mode. It is further found that since the relevant information of the first data packet can affect the reliability of air interface transmission, the first transmission mode is related to the relevant information of the first data packet. Therefore, this application proposes that the first transmission mode of the first data packet can be determined based on first information. The first information can be used to indicate the relevant information of the first data packet. As one implementation, the first information can be prediction information. The first information can be used to indicate at least one of the following: a first predicted size corresponding to the first data packet, a first predicted confidence level corresponding to the first data packet, a first predicted confidence interval corresponding to the first data packet, a first predicted priority corresponding to the first data packet, a first predicted arrival time corresponding to the first data packet, a second predicted confidence level corresponding to the first data packet, a second predicted confidence interval corresponding to the first data packet, a first predicted service type corresponding to the first data packet, or a first predicted QoS requirement corresponding to the first data packet, etc. The first predicted size can be used to indicate the predicted size of the first data packet. The first predicted priority can be used to indicate the predicted transmission priority or scheduling priority of the first data packet. The first predicted arrival time can be used to indicate the predicted time of receiving the first data packet. The first predicted confidence level can be used to indicate the confidence level of the first predicted size. The first predicted confidence interval can be used to indicate the confidence interval of the first predicted size. The second prediction confidence level can be used to indicate the confidence level of the first predicted arrival time. The second prediction confidence interval can be used to indicate the confidence interval of the first predicted arrival time. The first predicted service type can be used to indicate the predicted service type of the first data packet. The first predicted QoS requirement can be used to indicate the predicted quality of service required for transmitting the first data packet.

[0221] Based on the above, this application proposes that before the base station receives the first data packet to be transmitted, it acquires first information about the first data packet and determines a first transmission method for transmitting the first data packet based on the first information. Upon receiving the first data packet, the first data packet can be transmitted according to the previously determined first transmission method. The first transmission method can be matched with the network status.

[0222] To facilitate a better understanding of the inventive concept of the embodiments of this application, further explanation will be provided below with reference to FIG3.

[0223] Figure 3 is a schematic diagram of the data transmission method provided in the embodiments of this application.

[0224] The base station can acquire the first information of the first data packet at a first moment. At a second moment, it determines the first transmission method of the first data packet based on the first information. At a third moment, it acquires the first data packet again. Therefore, the base station can transmit the first data packet according to the first transmission method. The second moment can be no earlier than the first moment, and the third moment can be later than the second moment.

[0225] Since the base station determines the first transmission method for transmitting the first data packet based on the first information corresponding to the first data packet before acquiring the first data packet, and the first information is used to indicate information related to the first data packet, the first transmission method determined based on the first information can be applied to the transmission of the first data packet. In this case, when the first data packet is acquired, the first data packet is transmitted according to the first transmission method previously determined to be applicable to the transmission of the first data packet, thereby improving the reliability of air interface transmission.

[0226] Next, we need to consider how the base station obtains the first information, which can be considered from two aspects: how the first information is generated and how the base station obtains the first information.

[0227] Regarding how to generate the first piece of information, it was found that it can be determined based on influence information and mapping relationships. Influence information can be used to indicate information that can affect the first piece of information. Mapping relationships can be used to indicate the relationship between influence information and the first piece of information. Therefore, it is necessary to determine the influence information and how to construct the mapping relationship between influence information and the first piece of information.

[0228] Regarding the determination of influencing information, it was found that the second piece of information is related to the first piece of information, or the fourth piece of information is related to the first piece of information.

[0229] Regarding the second information, it can have the same content as the first information, but a different representation. For example, the second information can indicate at least one of the following: a first predicted size corresponding to the first data packet, a first predicted confidence level corresponding to the first data packet, a first predicted confidence interval corresponding to the first data packet, a first predicted priority corresponding to the first data packet, a first predicted arrival time corresponding to the first data packet, a second predicted confidence level corresponding to the first data packet, a second predicted confidence interval corresponding to the first data packet, a first predicted service type corresponding to the first data packet, or a first predicted QoS requirement corresponding to the first data packet, etc. The second information can be encrypted. The first information can be unencrypted.

[0230] Regarding the fourth information, it can be used to indicate information related to the first information. The fourth information can be used to indicate at least one of the following: a first complexity and a first instruction. The first complexity can be used to indicate the complexity of the first screen. The first instruction can be used to indicate the instruction that triggers the first screen. The instruction can include user instructions. It should be noted that the fourth information can also be used to indicate other content, as long as it is related to the first information.

[0231] The first complexity and the first instruction involved in the fourth information are explained below.

[0232] Regarding the first complexity, when the first data packet is a video data packet or an image data packet, the complexity of the first frame can be evaluated from at least one of the following: texture complexity, color complexity, spatial complexity, first frequency domain complexity, first statistical feature complexity, first structural complexity, or deep learning feature complexity. Texture complexity can be used to indicate at least one of the following: image contrast or image homogeneity, etc. Color complexity can be used to indicate at least one of the following: color contrast, color histogram, or color entropy, etc. Spatial complexity can be used to indicate at least one of the following: edge density or gradient magnitude, etc. First frequency domain complexity can be used to indicate at least one of the following: proportion of high-frequency components, proportion of low-frequency components, or power spectral density, etc. First statistical feature complexity can be used to indicate at least one of the following: image entropy and image variance, etc. First structural complexity can be used to indicate at least one of the following: image fractal dimension or image self-similarity, etc.

[0233] Therefore, when the first data packet is a video data packet or an image data packet, the first complexity can be used to indicate at least one of the following: first image entropy, first image variance, first edge density, first gradient magnitude, first image contrast, first color contrast, first color histogram, first color entropy, or first fractal dimension.

[0234] When the first data packet is an audio data packet, the complexity of the first frame can be evaluated from at least one of the following: second frequency domain complexity, time domain complexity, dynamic range, second statistical feature complexity, musical complexity, or second structural complexity. Second frequency domain complexity can indicate at least one of the following: spectral density, spectral entropy, number of frequency components, uniformity of frequency distribution, variation of frequency components, phase information, or multi-scale characteristics. Time domain complexity can indicate at least one of the following: waveform complexity or zero crossover rate. Dynamic range can refer to the amplitude difference between the strongest and weakest parts of the audio. Second statistical feature complexity can indicate at least one of the following: audio entropy or audio variance. Musical complexity can indicate at least one of the following: melody complexity, harmonic complexity, rhythmic complexity, timbre complexity, arrangement complexity, or performance complexity. Second structural complexity can indicate at least one of the following: audio fractal dimension or audio self-similarity.

[0235] Therefore, in the case that the first data packet is an audio data packet, the first complexity can be used to indicate at least one of the following: first audio entropy, first spectral entropy, first dynamic range, or first audio variance.

[0236] In the case that the first data packet is a tactile data packet, the first complexity can be used to indicate at least one of the following: the fineness of the first surface texture, the roughness of the first surface texture, the hardness of the first material, the elastic modulus of the first material, the first vibration frequency, the first vibration amplitude, the intensity of the first force feedback, the direction of the first force feedback, the duration of the first contact, the area of ​​the first contact, the first tactile entropy, or the first tactile fractal dimension.

[0237] It should be noted that the method for determining the first complexity is not limited in the embodiments of this application. The first complexity can also be used to indicate other parameters related to the generation of the first data packet.

[0238] The above explains the first complexity involved in the fourth information. The following explains the first instruction involved in the fourth information.

[0239] Regarding the first instruction, the first instruction can be used to indicate at least one of the following: a first playback mode instruction or a first attitude control instruction. The first playback mode can be used to indicate at least one of the following: a first playback speed, a first fast forward, a first random playback, a first sequential playback, or a first loop playback, etc.

[0240] Furthermore, regarding the fourth and second pieces of information, it was found that since the fourth piece of information is related to the first piece of information, the second piece of information can be determined based on the fourth piece of information. Therefore, in addition to directly establishing a mapping relationship between the fourth piece of information and the first piece of information, a mapping relationship can also be indirectly established between the fourth piece of information and the first piece of information through the second piece of information.

[0241] Regarding the determination of the influencing information, it was also found that the third information can be related to the first information. The third information can be used to indicate at least one of the following: first channel information or first available resources. The first channel information can be used to indicate information related to the first channel. The first channel can be used to transmit the first data packet. The first available resources can be used to indicate available resources on the second device side.

[0242] The first channel information and the first available resource are explained below.

[0243] Regarding the first channel information, the first channel information can be used to indicate at least one of the following: first channel state information, first channel time-varying information, first channel frequency response information, first signal-to-noise ratio, first signal-to-dryness ratio, or first path loss model parameters, etc.

[0244] The first channel state information can be used to indicate at least one of the following: first channel matrix, first channel quality indicator, first rank indicator, first precoding type indicator, or first precoding matrix indicator, etc.

[0245] The time-varying information of the first channel can be used to indicate information about how the first channel changes over time. For example, the time-varying information of the first channel may include at least one of the following: first Doppler frequency shift or first Doppler spread, etc.

[0246] The first channel frequency response information can be used to indicate the first channel's response to signals of different frequencies.

[0247] The first path loss model parameters can be used to indicate signal strength attenuation due to distance and environment during signal transmission. For example, the first path loss model parameters can indicate at least one of the following: a first path loss exponent, a first multipath fading parameter, or a first environmental characteristic parameter. The first path loss exponent can be used to indicate the rate at which signal strength attenuates with increasing distance. The first multipath fading parameter can be used to indicate the change in signal strength due to multipath transmission. The first environmental characteristic parameter can be used to indicate at least one of the following: a first terrain type, a first vegetation density, or a first building density.

[0248] The first available resource can be used to indicate at least one of the following: a first time-frequency resource, first network congestion information, first spectrum resource, first time resource, first power resource, first antenna resource, first code resource, first computing resource, first storage resource, first network bandwidth, first physical resource block, or first channel resource, etc. The first network congestion information can be used to determine the first predicted arrival time corresponding to the first data packet. The first time-frequency resource can be organized into resource blocks. Each resource block has its own size in the time and frequency domains.

[0249] It should be noted that the method for determining the third information is not limited in the embodiments of this application.

[0250] Regarding how to construct the mapping relationship between influencing information and primary information, it was found that a construction method based on a neural network model or a construction method based on a traditional mathematical model can be adopted.

[0251] To facilitate understanding, Figure 4 below will be used to explain the determination of the influence information mentioned above and how to construct the mapping relationship between the influence information and the first information.

[0252] Figure 4 is a schematic diagram of the principle of the method for generating first information provided in the embodiments of this application.

[0253] As shown in Figure 4, the influence information can be input into a mapping relationship to obtain the first information. The mapping relationship refers to the relationship between the influence information and the first information. The mapping relationship can be constructed based on a neural network model or a traditional mathematical model. The influence information may include a fourth piece of information. Optionally, the influence information may include both the fourth and third pieces of information. Optionally, the influence information may include a second piece of information. Optionally, the influence information may include both the second and third pieces of information. Optionally, the second information may be determined based on the fourth information. The second information may have the same content as the first information, but a different representation.

[0254] The following section explains how to construct the mapping relationship between influencing information and primary information.

[0255] Regarding the construction method based on traditional mathematical models, one implementation approach is to use a linear function as the traditional mathematical model. The parameters of the linear function can be determined using the first training sample. For example, the first training sample may include the fifth and sixth information.

[0256] Regarding the construction method based on neural network models, it was found that it can be achieved through either a construction method based on a single model or a construction method based on a distributed model. This will be explained below with reference to Figures 5A and 5B.

[0257] Figure 5A is a schematic diagram illustrating the principle of the construction method based on the neural network model provided in the embodiments of this application.

[0258] As shown in Figure 5A, the first approach is based on a single model. For example, a third neural network model is used to construct the relationship between the influencing information and the first information.

[0259] The second approach is based on a distributed model. For example, a first neural network model and a second neural network model can be used to construct a mapping relationship between the influencing information and the first information.

[0260] The input information of the third neural network model can be the input information of the second neural network model, i.e., the first input information. The output information of the third neural network model can be the output information of the first neural network model, i.e., the first output information.

[0261] The input information of the first neural network model may include the first output information of the second neural network model, or the input information of the first neural network model may include the first output information of the second neural network model and other input information. The first output information of the second neural network model may be the interaction information between the first neural network model and the second neural network model.

[0262] Furthermore, regarding the construction method based on the distributed model, it was found that it can be implemented in the following two ways.

[0263] The first approach is based on model segmentation. For example, the first and second neural network models can be obtained by segmenting the third neural network model. The first neural network model can be a first part of the third neural network model. The second neural network model can be a second part of the third neural network model. This application does not limit the model segmentation method in its embodiments.

[0264] The second approach is based on independent models. For example, both the first and second neural network models are independent neural network models.

[0265] It should be noted that the model structures of the first neural network model, the second neural network model, and the third neural network model can be configured according to actual business needs in this application embodiment. There is no limitation here, as long as the mapping relationship between the influence information and the first information can be constructed.

[0266] For example, a neural network model can be constructed based on at least one model structure and the connections between different model structures. The model structure can include at least one model substructure and the connections between different model substructures. The model structure can be obtained by connecting at least one model substructure based on the connections between different model substructures. The model substructure can be a structure derived from at least one operational layer. Therefore, the model structure can be a structure obtained by connecting at least one model substructure from each of multiple operational layers based on the connections between different model substructures. For example, at least one operational layer can include an input layer, a hidden layer, and an output layer, etc.

[0267] As one implementation approach, the third neural network model can include a feedforward neural network model. Optionally, the feedforward neural network model can include a backpropagation (BP) neural network model. The number of hidden layers in the BP neural network model can be configured according to actual business needs and is not limited here.

[0268] To facilitate understanding of the model-based segmentation method described above, the following example illustrates this: "A third neural network model may include an input layer, hidden layers, and an output layer. The hidden layers may include a first hidden layer and a second hidden layer. The input layer may include multiple neurons. The first hidden layer may include multiple neurons. The second hidden layer may include multiple neurons. The output layer may include multiple neurons."

[0269] Figure 5B is a schematic diagram of the model segmentation method provided in the embodiment of this application.

[0270] As shown in Figure 5B, the third neural network model can be divided into a first neural network model and a second neural network model. The first neural network model may include a second hidden layer and an output layer. The second neural network model may include an input layer and a first hidden layer.

[0271] Next, we need to consider how to train the neural network model and how to deploy it.

[0272] Training a neural network model involves constructing training samples and developing training strategies. These will be explained separately below.

[0273] Regarding the construction of training samples, as mentioned above, since it is necessary to establish a mapping relationship between influence information and primary information, the training samples are related to both influence information and primary information. Furthermore, it is also necessary to combine the construction methods mentioned above to form training samples corresponding to each of the various construction methods.

[0274] For a model-based construction approach, the second training sample may include the fifth and sixth information. Optionally, the second training sample may include the fifth, seventh, and sixth information.

[0275] For the distributed model-based construction approach, when using a model-segmentation method, the third training sample can be the second training sample. When using an independent model-based approach, the third training sample used to train the first neural network model can include the eighth and sixth information. Optionally, the third training sample used to train the first neural network model can include the eighth, seventh, and sixth information. The third training sample used to train the second neural network model can include the fifth and eighth information. Optionally, the third training sample used to train the second neural network model can include the fifth, seventh, and eighth information.

[0276] The fifth piece of information can be similar to the fourth piece of information described above. The fifth piece of information can be used to indicate information related to the sixth piece of information. The fifth piece of information can be used to indicate at least one of the following: second complexity and second instruction. The second instruction can be used to indicate the instruction that triggers the second screen. The second complexity can be used to indicate the complexity of the second screen. It should be noted that the fifth piece of information can also be used to indicate other content, as long as it is related to the sixth piece of information.

[0277] When the second data packet is a video data packet or an image data packet, the second complexity can be used to indicate at least one of the following: second image entropy, second image variance, second edge density, second gradient magnitude, second image contrast, second color contrast, second color histogram, second color entropy, or second fractal dimension. When the second data packet is an audio data packet, the second complexity can be used to indicate at least one of the following: second audio entropy, second spectral entropy, second dynamic range, or second audio variance. When the second data packet is a haptic data packet, the second complexity can be used to indicate at least one of the following: second surface texture fineness, second surface texture roughness, second material hardness, second material elastic modulus, second vibration frequency, second vibration amplitude, second force feedback intensity, second force feedback direction, second contact duration, second contact area, second haptic entropy, or second haptic fractal dimension.

[0278] The second instruction is used to indicate at least one of the following: a second playback mode instruction or a second attitude control instruction. The second playback mode can be used to indicate at least one of the following: a second playback speed, a second fast forward, a second random playback, a second sequential playback, or a second loop playback, etc.

[0279] The seventh information can be similar to the third information described above. The seventh information can be used to indicate at least one of the following: second channel information or second available resources. Second channel information can be used to indicate information related to the second channel. The second channel can be used to transmit a second data packet. Second channel information can be used to indicate at least one of the following: second channel state information, second channel time-varying information, second channel frequency response information, second signal-to-noise ratio, second signal-to-dryness ratio, or second path loss model parameters, etc. Second available resources can be used to indicate available resources on the second device side or available resources on other second device sides. Other second device sides can refer to second device sides other than the second device side itself. For example, the second device side can be RAN node 130-1 in Figure 1. Other second device sides can be RAN node 130-N in Figure 1.

[0280] The second channel state information can be used to indicate at least one of the following: a second channel matrix, a second channel quality indicator, a second rank indicator, a second precoding type indicator, or a second precoding matrix indicator, etc. The second channel time-varying information can include at least one of the following: a second Doppler frequency shift or a second Doppler spread, etc. The second channel frequency response information can be used to indicate the response information of the second channel to signals of different frequencies. The second path loss model parameters can be used to indicate at least one of the following: a second path loss exponent, a second multipath fading parameter, or a second environmental characteristic parameter, etc. The second environmental characteristic parameters can be used to indicate at least one of the following: a second terrain type, a second vegetation density, or a second building density, etc.

[0281] The second available resource can be used to indicate at least one of the following: a second time-frequency resource, second network congestion information, second spectrum resource, second time resource, second power resource, second antenna resource, second code resource, second computing resource, second storage resource, second network bandwidth, second physical resource block, or second channel resource, etc. The second network congestion information can be used to determine the third predicted arrival time corresponding to the second data packet.

[0282] The sixth information can be similar to the first information described above. The sixth information can be used as tag information. The sixth information can be used to indicate relevant information about the second data packet. As one implementation, the sixth information can be used to indicate at least one of the following: a first actual size corresponding to the second data packet, a first actual confidence level corresponding to the second data packet, a first actual confidence interval corresponding to the second data packet, a first actual priority corresponding to the second data packet, a first actual arrival time corresponding to the second data packet, a second actual confidence level corresponding to the second data packet, a second actual confidence interval corresponding to the second data packet, a first actual service type corresponding to the second data packet, or a first actual QoS requirement corresponding to the second data packet, etc. The first actual size can be used to indicate the actual size of the second data packet. The first actual priority can be used to indicate the transmission priority or scheduling priority of the actual second data packet. The first actual arrival time can be used to indicate the actual time when the second data packet was received. The first actual confidence level can be used to indicate the confidence level of the first actual size. The first actual confidence interval can be used to indicate the confidence interval of the first actual size. The second actual confidence level can be used to indicate the confidence level of the first actual arrival time. The second actual confidence interval can be used to indicate the confidence interval of the first actual arrival time. The first actual service type can be used to indicate the actual service type of the second data packet. The first actual QoS requirement can be used to indicate the actual quality of service required for transmitting the second data packet.

[0283] The eighth piece of information can be similar to the sixth piece of information described above. The eighth piece of information can be used to indicate information related to the sixth piece of information. As one implementation, the eighth piece of information can have the same content as the sixth piece of information, but a different representation. For example, the eighth piece of information can be used to indicate at least one of the following: the first actual size corresponding to the second data packet, the first actual confidence level corresponding to the second data packet, the first actual confidence interval corresponding to the second data packet, the first actual priority corresponding to the second data packet, the first actual arrival time corresponding to the second data packet, the second actual confidence level corresponding to the second data packet, the second actual confidence interval corresponding to the second data packet, the first actual service type corresponding to the second data packet, or the first actual QoS requirement corresponding to the second data packet, etc. The eighth piece of information can be encrypted. The sixth piece of information can be unencrypted.

[0284] When constructing a training strategy, it is necessary to consider the training process, the entity executing the training process, and the method of obtaining training samples. These will be explained in detail below with reference to the accompanying diagrams.

[0285] The model training process in Figure 5A will be explained below with reference to Figures 6A, 6B, and 6C.

[0286] Figure 6A is a schematic diagram illustrating the principle of a neural network model training method provided in an embodiment of this application. This method can be applied to the model-based construction method shown in Figure 5A, as well as the model segmentation method in the distributed model construction method.

[0287] As shown in Figure 6A, for the construction method based on a single model, the second training sample may include the fifth and sixth information. Optionally, the second training sample may include the fifth, seventh, and sixth information.

[0288] The third neural network model is trained using the second training samples. For example, the fifth piece of information is input into the third neural network model to obtain the ninth piece of information. Optionally, the fifth and seventh pieces of information are input into the third neural network model to obtain the ninth piece of information. The ninth and sixth pieces of information are input into the loss function to obtain the first loss function value. The model parameters of the third neural network model are adjusted according to the first loss function value to obtain the trained third neural network model.

[0289] The ninth piece of information can be used to indicate relevant information about the second data packet. As one implementation, the ninth piece of information can be used to indicate at least one of the following: a second predicted size corresponding to the second data packet, a third predicted confidence level corresponding to the second data packet, a third predicted confidence interval corresponding to the second data packet, a second predicted priority corresponding to the second data packet, a second predicted arrival time corresponding to the second data packet, a fourth predicted confidence level corresponding to the second data packet, a fourth predicted confidence interval corresponding to the second data packet, a second predicted service type corresponding to the second data packet, or a second predicted QoS requirement corresponding to the second data packet, etc. The second predicted size can be used to indicate the predicted size of the second data packet. The second predicted priority can be used to indicate the predicted transmission priority or scheduling priority of the second data packet. The second predicted arrival time can be used to indicate the predicted time when the second data packet is received. The third predicted confidence level can be used to indicate the confidence level of the second predicted size. The third predicted confidence interval can be used to indicate the confidence interval of the second predicted size. The fourth predicted confidence level can be used to indicate the confidence level of the second predicted arrival time. The fourth predicted confidence interval can be used to indicate the confidence interval of the second predicted arrival time. The second predicted service type can be used to indicate the predicted service type of the second data packet. The second predicted QoS requirement can be used to indicate the predicted quality of service required for transmitting the second data packet.

[0290] Regarding the construction method based on the distributed model,

[0291] Regarding the model-based segmentation method, the third training sample can be the second training sample. Correspondingly, the training method described above can be used to train the third neural network model. The first neural network model can be the first part of the third neural network model. The second neural network model can be the second part of the third neural network model.

[0292] It should be noted that the first input information in Figure 5A may include the fifth information. Optionally, the first input information in Figure 5A may include both the fifth and seventh information. The first output information in Figure 5A may include the ninth information.

[0293] Figure 6B is a schematic diagram illustrating the principle of another neural network model training method provided in an embodiment of this application. This method can be applied to the independent model approach in the distributed model construction method shown in Figure 5A, for training the first neural network model.

[0294] The third training sample used to train the first neural network model may include the eighth information and the sixth information. Optionally, the third training sample used to train the first neural network model may include the eighth information, the seventh information, and the sixth information.

[0295] The first neural network model is trained using the third training sample. For example, the eighth piece of information is input into the first neural network model to obtain the tenth piece of information. Optionally, the eighth and seventh pieces of information are input into the first neural network model to obtain the tenth piece of information. The tenth and sixth pieces of information are input into the loss function to obtain the second loss function value. The model parameters of the first neural network model are adjusted according to the second loss function value to obtain the trained first neural network model.

[0296] The tenth piece of information can be used to indicate at least one of the following: the third predicted size corresponding to the second data packet, the fifth predicted confidence level corresponding to the second data packet, the fifth predicted confidence interval corresponding to the second data packet, the third predicted priority corresponding to the second data packet, the third predicted arrival time corresponding to the second data packet, the sixth predicted confidence interval corresponding to the second data packet, the third predicted service type corresponding to the second data packet, or the third predicted QoS requirement corresponding to the second data packet, etc. The tenth piece of information can be unencrypted information. The third predicted size can be used to indicate the predicted size of the second data packet. The third predicted priority can be used to indicate the predicted transmission priority or scheduling priority of the second data packet. The third predicted arrival time can be used to indicate the predicted time when the second data packet is received. The fifth predicted confidence level can be used to indicate the confidence level of the third predicted size. The fifth predicted confidence interval can be used to indicate the confidence interval of the third predicted size. The sixth predicted confidence level can be used to indicate the confidence level of the third predicted arrival time. The sixth predicted confidence interval can be used to indicate the confidence interval of the third predicted arrival time. The third predicted service type can be used to indicate the predicted service type of the second data packet. The third predicted QoS requirement can be used to indicate the predicted quality of service required for transmitting the second data packet.

[0297] Figure 6C is a schematic diagram illustrating the principle of another neural network model training method provided in an embodiment of this application. This method can be applied to the independent model approach in the distributed model construction method shown in Figure 5A, for training the second neural network model.

[0298] The third training sample used to train the second neural network model may include the fifth and eighth information. Optionally, the third training sample used to train the second neural network model may include the fifth, seventh, and eighth information.

[0299] The second neural network model is trained using the third training sample. For example, the fifth piece of information is input into the second neural network model to obtain the eleventh piece of information. Optionally, the fifth and seventh pieces of information are input into the second neural network model to obtain the eleventh piece of information. The ninth and eighth pieces of information are input into the loss function to obtain the third loss function value. The model parameters of the second neural network model are adjusted based on the third loss function value to obtain the trained second neural network model.

[0300] The eleventh piece of information can be used to indicate at least one of the following: the fourth predicted size corresponding to the second data packet, the seventh predicted confidence level corresponding to the second data packet, the seventh predicted confidence interval corresponding to the second data packet, the fourth predicted priority corresponding to the second data packet, the fourth predicted arrival time corresponding to the second data packet, the eighth predicted confidence level corresponding to the second data packet, the eighth predicted confidence interval corresponding to the second data packet, the fourth predicted type corresponding to the second data packet, or the fourth predicted QoS requirement corresponding to the second data packet, etc. The eleventh piece of information can be encrypted. The fourth predicted size can be used to indicate the predicted size of the second data packet. The fourth predicted priority can be used to indicate the predicted transmission priority or scheduling priority of the second data packet. The fourth predicted arrival time can be used to indicate the predicted time when the second data packet is received. The seventh predicted confidence level can be used to indicate the confidence level of the fourth predicted size. The seventh predicted confidence interval can be used to indicate the confidence interval of the fourth predicted size. The eighth predicted confidence level can be used to indicate the confidence level of the fourth predicted arrival time. The eighth predicted confidence interval can be used to indicate the confidence interval of the fourth predicted arrival time. The fourth predicted service type can be used to indicate the predicted service type of the second data packet. The fourth predicted QoS requirement can be used to indicate the predicted quality of service required for transmitting the second data packet.

[0301] It should be noted that the loss function in this application embodiment can be configured according to actual business needs, and is not limited thereto. For example, the loss function may include at least one of the following: mean squared error function, mean absolute error function, or cross-entropy loss function, etc.

[0302] Regarding the execution entity of the training process, it is found that the embodiments of this application involve a first device side and a second device side. The second device side can be an access side, and the first device side can be a service side or a terminal side. Both the first and second device sides have model training capabilities. Therefore, the execution entity of the training process can include the first device side and / or the second device side. For example, the training process for a third neural network model can be executed by either the first or the second device side. The training process for a first neural network model can be executed by either the first or the second device side. The training process for a second neural network model can be executed by either the first or the second device side. The execution entity of the training process in Figures 6A, 6B, and 6C will be described below with reference to Figure 7.

[0303] Figure 7 is a schematic diagram of the execution entity of the training process provided in the embodiment of this application.

[0304] As shown in Figure 7, in the first approach, the third neural network model can be trained by the second device. In the second approach, the third neural network model can be trained by the first device. In the third approach, both the first and second neural network models can be trained by the second device. In the fourth approach, both the first and second neural network models can be trained by the first device. In the fifth approach, both the first and second neural network models can be trained by the second device. As one implementation, the first device can be a server or a terminal device. The second device can be a base station.

[0305] Regarding the method of obtaining training samples, it was found that it needs to be determined based on the provider of the training samples and the entity executing the training process. The provider of the fifth and sixth information involved in the second training samples, as well as the eighth information involved in the third training samples, can be the first device side. The provider of the second available resources involved in the second and third training samples can be the second device side.

[0306] Since the fifth and sixth information used to train the third neural network model, or the fifth and sixth information used to train the first neural network model, or the eighth and sixth information used to train the first neural network model, are provided by the first device, when the third neural network model is trained by the second device or the first neural network model is trained by the second device, the second device needs to obtain the fifth and sixth information, or the eighth and sixth information, from the first device. This can be achieved by the second device sending a first signaling instruction to the first device. The first signaling instruction can be used to instruct the first device to send the fifth and sixth information to the second device via a first method, or to instruct the first device to send the eighth and sixth information to the second device via a first method. The first method can be used to carry the fifth and sixth information. Alternatively, the first method can be used to carry the eighth and sixth information. Furthermore, the first signaling instruction can also be used to indicate the size of the fifth and sixth information. Alternatively, the first signaling instruction can also be used to indicate the size of the eighth and sixth information.

[0307] Since the first device side can be either a service side or a terminal side, and the second device side can be an access side, the first method between the access side and the service side differs from the first method between the access side and the terminal side. Therefore, the first method also needs to be configured based on the interaction object with the access side. The interaction object can be either a service side or a terminal side. As one implementation, the service side can be a server. The terminal side can be a terminal device. The access side can be a base station.

[0308] As one implementation, when the first device is the service side and the second device is the access side, the first approach may include at least one of the following: a GTP-U header or a first protocol between the access side and the service side. The first protocol may include fields for indicating fifth information and sixth information. Alternatively, the first protocol may include fields for indicating eighth information and sixth information.

[0309] When the first device side is the terminal side and the second device side is the access side, the first method may include at least one of the following: a MAC CE or a second protocol between the access side and the terminal side. The second protocol may include fields for indicating fifth information and sixth information. Alternatively, the second protocol may include fields for indicating eighth information and sixth information. The first signaling may also be used to indicate a transmission period. The transmission period may be used to indicate the period at which the terminal side sends fifth and sixth information to the access side. Alternatively, the transmission period may be used to indicate the period at which the terminal side sends eighth and sixth information to the access side. The transmission period may be configured using RRC or triggered via a MAC CE. Triggering may include a single trigger or a periodic trigger following a single trigger.

[0310] For ease of understanding, the first or second agreement will be explained below with reference to Figure 8.

[0311] Figure 8 is a schematic diagram of a first protocol or an example of a first protocol provided in an embodiment of this application.

[0312] As shown in Figure 8, the first or second protocol may include fields for indicating fifth information and fields for indicating sixth information. For example, the fifth information may indicate the second complexity and the second instruction. The sixth information may indicate the actual size, actual priority, and actual arrival time corresponding to the second data packet. Furthermore, the second complexity can be carried using X bits. The second instruction can be carried using Y bits. The actual size can be carried using Z bits. The actual priority can be carried using U bits. The actual arrival time can be carried using W bits. As one implementation, (fifth information) → (sixth information) can form a map relationship. For example, (second complexity, second instruction) → (actual size, actual priority, actual arrival time).

[0313] It should be noted that Figure 8 is merely an exemplary example and does not constitute a limitation on the embodiments of this application.

[0314] Regarding the deployment of neural network models, it was found that the first neural network model can be deployed on the second device, the second neural network model can be deployed on the first device, and the third neural network model can be deployed on either the second or first device. When the execution entity and the deployment entity differ during the training process, the execution entity needs to send the model parameters of the corresponding neural network model to the deployment entity so that the deployment entity can deploy the appropriate neural network model.

[0315] As one implementation, when the first neural network model is trained by the first device and deployed on the second device, the first device can send the model parameters of the first neural network model to the second device, and the second device can deploy the first neural network model according to the model parameters of the first neural network model.

[0316] When the second neural network model is trained on the second device side and deployed on the first device side, the second device side can send the model parameters of the second neural network model to the first device side, and the first device side can deploy the second neural network model according to the model parameters of the second neural network model.

[0317] When the third neural network model is trained on the first device side and deployed on the second device side, the first device side can send the model parameters of the third neural network model to the second device side, and the second device side can deploy the third neural network model according to the model parameters of the third neural network model.

[0318] When the third neural network model is trained by the second device and deployed on the first device, the second device can send the model parameters of the third neural network model to the first device, and the first device can deploy the third neural network model according to the model parameters.

[0319] Next, it is necessary to consider the scenario where the first and second neural network models are deployed on different entities. Since the cooperation between the first and second neural network models is required to obtain the first information, it is necessary to construct interaction information between the first and second neural network models. For example, at least one of the following needs to be determined: the meaning of the interaction information, the size of the interaction information, or a second method for carrying the interaction information. The form of the interaction information can be configured according to actual business needs and is not limited here. For example, the interaction information can be interaction token information. The interaction token information can include at least one interaction token (referred to as "token").

[0320] As one implementation, if the first neural network model is deployed on the second device side and the second neural network model is deployed on the first device side, and if both the first and second neural network models are trained on the second device side, then the second device side can send a second signaling message to the first device side via a third method. Alternatively, the second signaling message can be used to indicate the second method and the interaction information between the second device side and the first device side. Or, the second signaling message can be used to indicate the second method, the interaction information, and the magnitude of the interaction information. The second information can be interaction information.

[0321] Since the first device side can be either a service side or a terminal side, and the second device side can be an access side, the second method between the access side and the service side differs from the second method between the access side and the terminal side. Similarly, the third method between the access side and the service side differs from the third method between the access side and the terminal side. Therefore, the second and third methods need to be configured based on the interaction object with the access side. The interaction object can be either a service side or a terminal side. As one implementation, the service side can be a server. The terminal side can be a terminal device. The access side can be a base station.

[0322] Regarding the second approach, where the first device is the service side and the second device is the access side, the second approach may include at least one of the following: a GTP-U header or a third protocol between the access side and the service side. The third protocol may include fields for indicating interaction information.

[0323] When the first device side is the terminal side and the second device side is the access side, the second method may include at least one of the following: UCI, UAI, or a fourth protocol between the access side and the terminal side. The fourth protocol may include fields for indicating interaction information.

[0324] Regarding the third approach, where the first device is the serving side and the second device is the access side, the third approach may include at least one of the following: a GTP-U header or a fifth protocol between the access side and the serving side. The fifth protocol may include fields for indicating the second signaling.

[0325] When the first device side is the terminal side and the second device side is the access side, the third method may include at least one of the following: MAC CE, DCI, or a sixth protocol between the access side and the terminal side. The sixth protocol may include fields for indicating the second signaling. The access side can obtain inference data (e.g., application layer quality of service) sent by the terminal side via DCI, as well as inference data indicating the past T periods. T can be an integer greater than or equal to 1. Furthermore, the terminal side can also send inference data to the access side via MAC CE.

[0326] Furthermore, to enhance information security, the exchanged information can be configured as encrypted. The size of the exchanged information can be optimized to minimize transmission.

[0327] For ease of understanding, the fifth or sixth protocol will be explained below with reference to Figure 9.

[0328] Figure 9 is a schematic diagram of an example of the fifth or sixth protocol provided in the embodiments of this application.

[0329] As shown in Figure 9, the fifth or sixth protocol may include fields for indicating interaction information. As one implementation, the interaction information can be represented by interaction tokens. The interaction information can be used to indicate the predicted size (i.e., the first token), predicted confidence level (i.e., the second token), predicted confidence interval (i.e., the third token), predicted priority (i.e., the fourth token), and predicted arrival time (i.e., the fifth token) corresponding to the first data packet. Furthermore, the predicted size can be defined as being carried in Mbits. The predicted confidence level can be carried in Nbits. The predicted confidence interval can be carried in Obits. The predicted priority can be carried in Pbits. The predicted arrival time can be carried in Qbits.

[0330] It should be noted that Figure 9 is merely an exemplary example and does not constitute a limitation on the embodiments of this application.

[0331] For ease of understanding, the following description, in conjunction with Figure 10, uses a server or terminal device as the first device side and a base station as the second device side to illustrate how to obtain the first information using the neural network model described in the embodiments of this application.

[0332] Figure 10 is a schematic diagram of the principle of obtaining first information based on a neural network model according to an embodiment of this application.

[0333] As shown in Figure 10, when the first neural network model is deployed at the base station (e.g., the base station MAC layer) and the second neural network model is deployed at the server (e.g., the server application layer), the server can input the fourth information into the second neural network model to obtain the second information. The second information can be interactive information. The base station can obtain the first information based on the second information and the first neural network model. For example, the base station can input the second information into the first neural network model to obtain the first information, and then obtain the meaning of the first information based on the meaning of the interactive information. Optionally, the base station can obtain the first information based on the first neural network model, using both the second and third information. For example, the base station can input the second and third information into the first neural network model to obtain the first information, and then obtain the meaning of the first information based on the meaning of the interactive information.

[0334] When the first neural network model is deployed at the base station (e.g., the base station MAC layer) and the second neural network model is deployed at the terminal device (e.g., the terminal device application layer), the terminal device can input the fourth information into the second neural network model to obtain the second information. The base station can obtain the first information based on the first neural network model and the second information. Optionally, the base station can obtain the first information based on the first neural network model and the second and third information.

[0335] When the third neural network model is deployed at the base station (e.g., the base station MAC layer), the base station can input the fourth information into the third neural network model to obtain the first information. Optionally, the base station can input both the fourth and third information into the third neural network model to obtain the first information.

[0336] When the third neural network model is deployed on a server (e.g., the server application layer), the server can input the fourth information into the third neural network model to obtain the first information. Optionally, the server can input both the fourth and third information into the third neural network model to obtain the first information.

[0337] When the third neural network model is deployed on a terminal device (e.g., the terminal device application layer), the terminal device can input the fourth information into the third neural network model to obtain the first information. Optionally, the terminal device can input both the fourth and third information into the third neural network model to obtain the first information.

[0338] By establishing a source graph between the second device and the first device, the second device can transmit data packets according to a pre-determined first transmission method based on an advance awareness of the future size of the first data packet, thereby improving the reliability of air interface transmission and enhancing user experience. In this scenario, the number of users that the second device can serve is increased under limited air interface resources, thus improving air interface resource utilization. The source graph may include at least one of the following: a first neural network model deployed on the second device, a first neural network model deployed on the first device, interaction information between the second and first devices, a second method carrying the interaction information, a fifth piece of information, a sixth piece of information, a seventh piece of information, or an eighth piece of information, etc.

[0339] The above describes the inventive concept of this application.

[0340] The data transmission method described in the embodiments of this application will now be described with reference to the accompanying drawings. FIG11 illustrates a data transmission method applied to a second device. FIG12-FIG.14 illustrate data transmission methods applied to a first device. FIG12 illustrates a data transmission method applied to a first device when the first device sends second information to the second device. FIG13 illustrates a data transmission method applied to a first device when the first device sends fourth information to the second device. FIG14 illustrates a data transmission method applied to a first device when the first device sends first information to the second device.

[0341] In one implementation, the second device can be an access side. The access side can be RAN node 130 in Figure 1. For example, RAN node 130 can be a base station. The base station can be a 4G, 5G, 5.5G, 6G, or a base station in a future communication system; this embodiment does not limit this.

[0342] In one implementation, the first device side can be either a service side or a terminal side. The service side can include a server. For example, the server can be server 110 in Figure 1. The terminal side can include a terminal device. The terminal device can be terminal device 140 in Figure 1. The server or terminal device can be a server or terminal device in a 4G, 5G, 5.5G, 6G, or future communication system, and this application embodiment does not limit this.

[0343] Figure 11 is a flowchart of a data transmission method provided in an embodiment of this application. This method can be applied to a second device.

[0344] As shown in Figure 11, the method may include S1110-S1120.

[0345] In S1110, the first information is obtained at the first moment.

[0346] According to embodiments of this application, the first information can be used to indicate relevant information about the first data packet. The first data packet can refer to a data packet to be transmitted by the second device. The first data packet can be a data packet for a multimedia transmission service. The multimedia transmission service can include at least one of the following: video transmission, image transmission, audio transmission, haptic transmission, or interactive media transmission, etc.

[0347] According to embodiments of this application, the first information can be prediction information. As one implementation, the first information can be used to indicate at least one of the following: a first prediction size corresponding to the first data packet, a first prediction confidence level corresponding to the first data packet, a first prediction confidence interval corresponding to the first data packet, a first prediction priority corresponding to the first data packet, a first prediction arrival time corresponding to the first data packet, a second prediction confidence level corresponding to the first data packet, a second prediction confidence interval corresponding to the first data packet, a first prediction type corresponding to the first data packet, or a first prediction QoS requirement corresponding to the first data packet, etc.

[0348] The following methods can be used to obtain the initial information.

[0349] In one approach, the second device can determine the first information based on the second information and the first neural network model. Optionally, the second device can determine the first information based on the second information and the third information and the first neural network model. The first neural network model can be pre-trained. It should be noted that explanations of the second and third information can be found in the relevant sections above and will not be repeated here.

[0350] Regarding how the second device can determine the first information based on the first neural network model and the second information, it can be achieved in the following way.

[0351] In one implementation, the second device inputs the second information into the first neural network model to obtain the first information. The meaning of the first information is determined based on the meaning of the interaction information. The second information can be interaction information.

[0352] Regarding how the second device can determine the first information based on the first neural network model and the second and third information, it can be achieved in the following way.

[0353] As one implementation method, the second device can input the second and third information into the first neural network model to obtain the first information. The meaning of the first information is then determined based on the meaning of the interaction information.

[0354] The first neural network model can be obtained in the following way.

[0355] As one implementation, the first neural network model can be the first part of the third neural network model. The third neural network model can be obtained in the following way.

[0356] As one implementation method, the second training samples include the fifth and sixth information. That is, the third neural network model can be trained on either the second device side or the first device side using the fifth and sixth information.

[0357] The second or first device side can use the fifth and sixth information to train the third neural network model in the following way.

[0358] The second or first device inputs the fifth information into the third neural network model to obtain the ninth information. The ninth information can be used to indicate relevant information about the second data packet. The second or first device inputs the ninth and sixth information into a loss function to obtain a first loss function value. The model parameters of the third neural network model are adjusted based on the first loss function value to obtain a trained third neural network model.

[0359] As another implementation, the second training sample includes the fifth, seventh, and sixth information. That is, the third neural network model can be trained on either the second or first device using the fifth, sixth, and seventh information.

[0360] The second or first device side can use the fifth, sixth, and seventh information to train the third neural network model in the following way.

[0361] The second or first device inputs the fifth and seventh information into the third neural network model to obtain the ninth information. The ninth and sixth information are then input into the loss function to obtain the first loss function value. The model parameters of the third neural network model are adjusted based on the first loss function value to obtain the trained third neural network model.

[0362] It should be noted that explanations of the fifth, sixth, seventh, and ninth pieces of information can be found in the corresponding sections above, and will not be repeated here.

[0363] As an alternative implementation, the first neural network model can be an independent neural network model.

[0364] The third training sample may include the eighth and sixth information. Optionally, the third training sample may include the eighth, seventh, and sixth information.

[0365] As one implementation method, the third training sample includes the eighth and sixth information. That is, the first neural network model can be trained on either the second or first device using the eighth and sixth information. For example, either the second or first device can input the eighth information into the first neural network model to obtain the tenth information. The second or first device can then input the tenth and sixth information into a loss function to obtain a second loss function value. The model parameters of the first neural network model are adjusted based on the second loss function value to obtain the trained first neural network model.

[0366] As another implementation, the third training sample includes the eighth, seventh, and sixth information. That is, the first neural network model can be trained on either the second or first device using the eighth, seventh, and sixth information. For example, the second or first device can input the eighth and seventh information into the first neural network model to obtain the tenth information. The tenth and sixth information are then input into a loss function to obtain the second loss function value. The model parameters of the first neural network model are adjusted based on the second loss function value to obtain the trained first neural network model.

[0367] In a second approach, the second device can receive fourth information from the first device. Based on the fourth information, the first information is determined. For example, the second device can input the fourth information into a third neural network model to obtain the first information. Optionally, the second device can determine the first information based on both the fourth and third information. For example, the second device can input the fourth and third information into a third neural network model to obtain the first information.

[0368] In a third approach, the second device can receive the first information from the first device. The first device can then determine the first information based on the fourth information. For example, the first device can input the fourth information into a third neural network model to obtain the first information. Optionally, the first device can determine the first information based on both the fourth and third information. For example, the first device can input the fourth and third information into a third neural network model to obtain the first information.

[0369] In the fourth method, the second device can input the fourth information into a linear function to obtain the first information. The linear function can be obtained by training a traditional mathematical model using the first training samples.

[0370] At time S1120, the first transmission mode of the first data packet is determined based on the first information.

[0371] According to embodiments of this application, the first transmission method can be used to transmit the first data packet when it is acquired at a third time. The second time is not earlier than the first time. The third time is later than the second time. For example, the first data packet can be transmitted at a fourth time according to the first transmission method. The fourth time may not be earlier than the third time.

[0372] According to embodiments of this application, the first transmission mode can be used to indicate at least one of the following: transmission time, resource size used, transmission order, frequency, time slot, transmit power, channel bandwidth, available resource blocks, modulation and coding scheme, or antenna selection scheme. Furthermore, the first transmission mode can also be used to indicate resource allocation information. Resource allocation information can refer to time resources and / or frequency resources allocated to the second device side, etc.

[0373] As one implementation, the second device can determine the transmission order of the first data packet based on at least one of the first prediction priority, the first prediction size, or the first prediction arrival time corresponding to the first data packet.

[0374] As an alternative implementation, the second device can determine the modulation and coding scheme based on the first predicted size corresponding to the first data packet. For example, if the first predicted size is greater than or equal to a threshold, a more efficient modulation and coding scheme can be used to increase the transmission rate. If the first predicted size is less than the threshold, a more reliable modulation and coding scheme can be used to reduce the bit error rate. The threshold can be configured according to actual business needs and is not limited here.

[0375] As an alternative implementation, the second device can determine the frequency and time slot based on the first predicted size corresponding to the first data packet. For example, if the first predicted size is greater than or equal to a threshold, more spectrum resources or longer time slots can be used to complete the transmission.

[0376] Since the second device determines a first transmission method for transmitting the first data packet based on the first information corresponding to the first data packet before acquiring the first data packet, and the first information is used to indicate information related to the first data packet, the first transmission method determined based on the first information can be applied to transmitting the first data packet. In this case, when the first data packet is acquired, the first data packet is transmitted according to the first transmission method previously determined to be applicable to transmitting the first data packet, thereby improving the reliability of air interface transmission.

[0377] As one implementation, the first transmission method can be used to transmit the first data packet according to the second transmission method when the first data packet is acquired at a third time. The second transmission method can be determined based on the first transmission method and the first data packet.

[0378] According to embodiments of this application, since the first information can be predicted information, there may be deviations. To further improve the reliability of air interface transmission, when the first data packet is acquired at a third time, the first transmission mode can be adjusted according to the twelfth information of the first data packet to obtain a second transmission mode, so that the first data packet can be transmitted according to the second transmission mode. The twelfth information can be used to indicate at least one of the following: a second actual size, a third actual confidence level, a third actual confidence interval, a second actual priority, a second actual arrival time, a fourth actual confidence level, a fourth actual confidence interval, a second actual type, or a second actual QoS requirement corresponding to the first data packet. The second actual size can be used to indicate the actual size of the first data packet. The second actual priority can be used to indicate the transmission priority or scheduling priority of the actual first data packet. The second actual arrival time can be used to indicate the actual time when the first data packet is received, i.e., the third time. The third actual confidence level can be used to indicate the confidence level of the second actual size. The third actual confidence interval can be used to indicate the confidence interval of the second actual size. The fourth actual confidence level can be used to indicate the confidence level of the second actual arrival time. The fourth actual confidence interval can be used to indicate the confidence interval of the second actual arrival time. The second actual service type can be used to indicate the actual service type of the first data packet. The second actual QoS requirement can be used to indicate the actual quality of service required to transmit the first data packet.

[0379] Since the first information can be predicted, and predictions may contain errors, while the first data packet acquired at the third moment is the actual data packet, adjusting the first transmission method based on the actual acquired first data packet yields the second transmission method, thus improving its accuracy. In this case, transmitting the first data packet according to the second transmission method further enhances the reliability of air interface transmission.

[0380] The following explains how the second device obtains the second information.

[0381] As one implementation, the second device can receive the second information sent by the first device through a second method.

[0382] The second method can refer to the method used to carry interactive information between the second device side and the first device side. The second information can be interactive information.

[0383] When the second device is the access side and the first device is the service side, the second method may include at least one of the following: a GTP-U header or a third protocol between the access side and the service side. The third protocol may include fields for indicating the second information.

[0384] When the second device side is the access side and the first device side is the terminal side, the second method may include at least one of the following: UCI, UAI, or a fourth protocol between the access side and the terminal side. The fourth protocol may include fields for indicating the second information.

[0385] The following section explains the interaction information between the second device side and the first device side, as well as the second method for carrying the interaction information.

[0386] When the first neural network model is trained by the second device, the second device can determine the interaction information between the second device and the first device, as well as a second method for carrying the interaction information. The second device can send a second signaling message to the first device, which can be used to indicate the interaction information between the second device and the first device, and the second method for carrying the interaction information.

[0387] When the first neural network model is trained by the second device, the second device sends a second signaling message to the first device. This second signaling message can be used to indicate a second mode and interaction information between the second device and the first device, or it can be used to indicate the second mode, interaction information, and the magnitude of the interaction information. The second information can be interaction information.

[0388] As one implementation method, the second device can send the second signaling to the first device through a third method.

[0389] When the first device is the service side and the second device is the access side, the third method may include at least one of the following: a GTP-U header or a fifth protocol between the access side and the service side. The fifth protocol may include fields for indicating the second signaling.

[0390] When the first device side is the terminal side and the second device side is the access side, the third method may include at least one of the following: MAC CE, Downlink Control Information (DCI), or a sixth protocol between the access side and the user equipment. The sixth protocol may include fields for indicating the second signaling.

[0391] The following explains how the second device obtains the fifth and sixth information.

[0392] In the case that the first neural network model can be trained by the second device, the fifth and sixth information can be obtained in the following way.

[0393] The second device can receive the fifth and sixth messages sent by the first device through the first method.

[0394] The first device side may obtain the fifth and sixth information to be sent to the second device side in the following manner, as well as the first method for carrying the fifth and sixth information.

[0395] The first method, the fifth information, and the sixth information can be indicated by the second device sending a first signaling message to the first device, which instructs the first device to send the fifth and sixth information to the second device via the first method. Optionally, the first signaling message can also be used to indicate the size of the fifth and sixth information. Optionally, when the first device is a terminal and the second device is an access device, the first signaling message can also be used to indicate the transmission period. The transmission period can be used to indicate the period during which the terminal sends the fifth and sixth information to the access device. The transmission period can be configured using RRC or triggered via MAC CE. The trigger can include a single trigger or a periodic trigger after a single trigger.

[0396] When the first device side is the service side and the second device side is the access side, the first method may include at least one of the following: a GTP-U header or a first protocol between the access side and the service side. The first protocol may include fields for indicating the fifth information and the sixth information.

[0397] When the first device side is the terminal side and the second device side is the access side, the first method may include at least one of the following: a MAC CE or a second protocol between the access side and the user equipment. The second protocol may include fields for indicating the fifth information and the sixth information.

[0398] By establishing a source graph between the second device and the first device, the second device can transmit data packets according to a pre-determined first transmission method based on an advance awareness of the future size of the first data packet, thereby improving the reliability of air interface transmission and enhancing user experience. In this scenario, the number of users that the second device can serve is increased under limited air interface resources, thus improving air interface resource utilization. The source graph may include at least one of the following: a first neural network model deployed on the second device, a first neural network model deployed on the first device, interaction information between the second and first devices, a second method carrying the interaction information, a fifth piece of information, a sixth piece of information, a seventh piece of information, or an eighth piece of information, etc.

[0399] Figure 12 is a flowchart of another data transmission method provided in an embodiment of this application. This method can be applied to the first device side.

[0400] As shown in Figure 12, the method may include S1210-S1220.

[0401] In S1210, obtain the second information.

[0402] According to an embodiment of this application, the second information may be determined by the first device based on the fourth information. It should be noted that explanations of the second and fourth information can be found in the corresponding sections above, and will not be repeated here.

[0403] The method for determining the second information based on the fourth information on the first device side can be implemented as follows.

[0404] In the first approach, the second information can be obtained by inputting the fourth information into the second neural network model from the first device side.

[0405] The second neural network model can be obtained in the following way.

[0406] As one implementation approach, the second neural network model can be the second part of the third neural network model. It should be noted that the training process for the third neural network model can be found in the relevant section above, and will not be repeated here.

[0407] As an alternative implementation, the second neural network model can be an independent neural network model.

[0408] The third training sample may include the fifth and eighth information. Optionally, the third training sample may include the fifth, seventh, and eighth information.

[0409] One implementation method involves using the third training samples, including the fifth and eighth information. That is, the second neural network model can be trained on either the second or first device using the fifth and eighth information. For example, the second or first device can input the fifth information into the second neural network model to obtain the eleventh information. The second or first device can then input the eleventh and eighth information into a loss function to obtain the third loss function value. The model parameters of the second neural network are adjusted based on the third loss function value to obtain the trained second neural network model.

[0410] As another implementation, the third training sample includes the fifth, seventh, and eighth information. That is, the second neural network model can be trained on either the second or first device using the fifth, seventh, and eighth information. For example, the second or first device can input the fifth and seventh information into the second neural network model to obtain the eleventh information. The eleventh and eighth information are then input into a loss function to obtain the third loss function value. The model parameters of the second neural network model are adjusted based on the third loss function value to obtain the trained second neural network model.

[0411] It should be noted that explanations of the fifth, seventh, eighth, and eleventh pieces of information can be found in the corresponding sections above, and will not be repeated here.

[0412] In S1220, the second information is sent to the second device side.

[0413] According to embodiments of this application, the second information can be used by the second device to acquire the first information at a first moment. The first information can be determined by the second device based on the second information and a first neural network model. Optionally, the first information can be determined based on the second information and a third information, using the first neural network model. At a second moment, a first transmission method for the first data packet is determined based on the first information. The first transmission method can be used to transmit the first data packet according to the first transmission method if the first data packet is acquired at a third moment. The second moment may not be earlier than the first moment. The third moment may be later than the second moment.

[0414] It should be noted that the explanations of the first information, the third information, and the first transmission method can be found in the corresponding sections above, and will not be repeated here.

[0415] Since the second device determines a first transmission method for transmitting the first data packet based on the first information corresponding to the first data packet before acquiring the first data packet, and the first information is used to indicate information related to the first data packet, the first transmission method determined based on the first information can be applied to transmitting the first data packet. In this case, when the first data packet is acquired, the first data packet is transmitted according to the first transmission method previously determined to be applicable to transmitting the first data packet, thereby improving the reliability of air interface transmission.

[0416] As one implementation, the first transmission method can be used to transmit the first data packet according to the second transmission method when the first data packet is acquired at a third time. The second transmission method can be determined based on the first transmission method and the first data packet.

[0417] According to embodiments of this application, since the first information can be predicted information, there may be deviations. To further improve the reliability of air interface transmission, when the first data packet is acquired at a third time, the first transmission mode can be adjusted according to the twelfth information of the first data packet to obtain a second transmission mode, so that the first data packet can be transmitted according to the second transmission mode. The twelfth information can be used to indicate at least one of the following: a second actual size, a third actual confidence level, a third actual confidence interval, a second actual priority, a second actual arrival time, a fourth actual confidence level, a fourth actual confidence interval, a second actual type, or a second actual QoS requirement corresponding to the first data packet, etc.

[0418] Since the first information can be predicted, and predictions may contain errors, while the first data packet acquired at the third moment is the actual data packet, adjusting the first transmission method based on the actual acquired first data packet yields the second transmission method, thus improving its accuracy. In this case, transmitting the first data packet according to the second transmission method further enhances the reliability of air interface transmission.

[0419] The following explains how the second device obtains the fifth and sixth information.

[0420] When the second neural network model is trained by the second device, the first device can receive a first signaling sent by the second device. The first signaling can instruct the first device to send fifth and sixth information to the second device via a first method. Optionally, the first signaling can also indicate the size of the fifth and sixth information. Optionally, when the first device is a terminal and the second device is an access side, the first signaling can also indicate a transmission period. The transmission period can indicate the period at which the terminal sends the fifth and sixth information to the access side. The transmission period can be configured using RRC or triggered via MAC CE. Triggering can include a single trigger or a single trigger followed by periodic triggering.

[0421] The first method will be explained below.

[0422] As one implementation, when the first device is the service side and the second device is the access side, the first approach may include at least one of the following: a GTP-U header or a first protocol between the access side and the service side. The first protocol may include fields for indicating the fifth information and the sixth information.

[0423] When the first device side is the terminal side and the second device side is the access side, the first method may include at least one of the following: MAC CE or a second protocol between the access side and the terminal side. The second protocol may include fields for indicating the fifth information and the sixth information.

[0424] The following explains how the first device sends the second information to the second device.

[0425] As one implementation method, the first device can send second information to the second device through a second method.

[0426] The following section explains the interaction information between the second device side and the first device side, as well as the second method for carrying the interaction information.

[0427] In one implementation, when the second neural network model is trained by the second device, the first device can receive second signaling from the second device. This second signaling can be used to indicate a second mode and interaction information between the second device and the first device. Alternatively, the second signaling can be used to indicate the second mode, interaction information, and the magnitude of the interaction information. The second information can be interaction information.

[0428] As one implementation, when the first device is the service side and the second device is the access side, the second approach may include at least one of the following: a GTP-U header or a third protocol between the access side and the service side. The third protocol may include fields for indicating the second information.

[0429] When the first device side is the terminal side and the second device side is the access side, the second method may include at least one of the following: UCI, UAI, or a fourth protocol between the access side and the terminal side. The fourth protocol may include fields for indicating the second information.

[0430] As one implementation, the first device can receive the second signaling sent by the second device via a third method.

[0431] The third method will be explained below.

[0432] When the first device is the service side and the second device is the access side, the third method may include at least one of the following: a GTP-U header or a fifth protocol between the access side and the service side. The fifth protocol includes fields for indicating the second signaling.

[0433] When the first device side is the terminal side and the second device side is the access side, the third method may include at least one of the following: MAC CE, DCI, or a sixth protocol between the access side and the terminal side. The sixth protocol includes fields for indicating the second signaling.

[0434] By establishing a source map between the second device and the first device, the second device can transmit data packets according to a pre-determined first transmission method, based on an advance awareness of the future size of the first data packet. This improves the reliability of air interface transmission and enhances the user experience. In this scenario, it increases the number of users the second device can serve, even with limited air interface resources, thereby improving air interface resource utilization.

[0435] Figure 13 is a flowchart of another data transmission method provided in an embodiment of this application. This method can be applied to the first device side.

[0436] As shown in Figure 13, the method may include S1310-S1320.

[0437] In S1310, obtain the fourth piece of information.

[0438] In S1320, the fourth message is sent to the second device side.

[0439] According to embodiments of this application, the fourth information can be used by the second device to acquire the first information at a first moment. The first information can be determined by the second device based on the fourth information. For example, the first information can be determined by the second device based on the fourth information using a third neural network model. Optionally, the first information can be determined by the second device based on the third neural network model using both the fourth and third information. At a second moment, a first transmission method for the first data packet is determined based on the first information. The first transmission method can be used to transmit the first data packet according to the first transmission method if the first data packet is acquired at a third moment. The second moment may not be earlier than the first moment. The third moment may be later than the second moment.

[0440] According to embodiments of this application, the first information can be used to indicate relevant information about the first data packet. The fourth information can be related to the first information.

[0441] It should be noted that the explanations of the fourth information, the first information, the third neural network model, and the first transmission method can be found in the corresponding sections above, and will not be repeated here.

[0442] Figure 14 is a flowchart of another data transmission method provided in an embodiment of this application. This method can be applied to the first device side.

[0443] As shown in Figure 14, the method may include S1410-S1420.

[0444] In S1410, obtain the first information.

[0445] According to embodiments of this application, the first information may be determined by the first device based on the fourth information. For example, the first information may be determined by the first device based on the fourth information and a third neural network model. Optionally, the first information may be determined by the first device based on the third neural network model and both the fourth and third information.

[0446] In S1420, the first information is sent to the second device side.

[0447] According to an embodiment of this application, the first information can be used by the second device to acquire the first information at a first moment. At a second moment, a first transmission method for the first data packet is determined based on the first information. The first transmission method can be used to transmit the first data packet according to the first transmission method if the first data packet is acquired at a third moment.

[0448] It should be noted that the explanations of the first information, the fourth information, the third information, the third neural network model, and the first transmission method can be found in the corresponding sections above, and will not be repeated here.

[0449] To facilitate a better understanding of how to obtain the first information in the data transmission method described in this application embodiment, the following explanation, in conjunction with the accompanying drawings, uses the second device side as a base station and the first device side as a server or terminal device as an example. However, this application embodiment does not limit the execution subject of the interactive illustration. Wherein:

[0450] Figures 15A-15C are applied to base stations and servers, with Figure 15A showing the training process. Figures 15B and 15C represent the inference process based on Figure 15A. In Figure 15A, the base station uses a model-based segmentation method to obtain the first and second neural network models (i.e., the third method in Figure 7 and the model-based segmentation method in Figure 5). Figure 15B uses the first and second neural network models from Figure 15A to obtain the first information. Figure 15C uses the third neural network model from Figure 15A to obtain the first information.

[0451] Figures 16A-16C are applied to base stations and servers. Figure 16A shows the training process. Figures 16B and 16C are the inference processes based on Figure 16A. In Figure 16A, the server uses a model-based segmentation method to obtain the first and second neural network models (i.e., the fourth method in Figure 7 and the model-based segmentation method in Figure 5). Figure 16B uses the first and second neural network models from Figure 16A to obtain the first information. Figure 16C uses the third neural network model from Figure 16A to obtain the first information.

[0452] Figures 17A-17C are applied to base stations and terminal equipment. Figure 17A shows the training process. Figures 17B and 17C are the inference processes based on Figure 17A. In Figure 17A, the base station uses a model-based segmentation method to obtain the first and second neural network models (i.e., the third method in Figure 7 and the model-based segmentation method in Figure 5). Figure 17B uses the first and second neural network models from Figure 17A to obtain the first information. Figure 17C uses the third neural network model from Figure 17A to obtain the first information.

[0453] Figures 18A-18C are applied to base stations and terminal devices. Figure 18A shows the training process. Figures 18B and 18C are the inference processes based on Figure 18A. In Figure 18A, the terminal device uses a model-based segmentation method to obtain the first and second neural network models (i.e., the fourth method in Figure 7 and the model-based segmentation method in Figure 5). Figure 18B uses the first and second neural network models from Figure 18A to obtain the first information. Figure 18C uses the third neural network model from Figure 18A to obtain the first information.

[0454] Figure 19 is applied to base stations and servers. Figure 19 shows the training process. The inference process corresponding to Figure 19 can be found in Figure 15B or 16B. In Figure 19, the base station uses an independent model-based approach to obtain the first neural network model and the second neural network model (i.e., the third approach in Figure 7 and the independent model-based approach in Figure 5).

[0455] Figure 20 is applied to the base station and server. Figure 20 shows the training process. The inference process corresponding to Figure 20 can be found in Figure 15B or 16B. In Figure 20, the server uses an independent model-based approach to obtain the first neural network model and the second neural network model (i.e., the fourth approach in Figure 7 and the independent model-based approach in Figure 5).

[0456] Figure 21 is applied to base stations and terminal devices. Figure 21 shows the training process. The inference process corresponding to Figure 21 can be found in Figure 17B or 18B. In Figure 21, the base station uses an independent model-based approach to obtain the first neural network model and the second neural network model (i.e., the third approach in Figure 7 and the independent model-based approach in Figure 5).

[0457] Figure 22 is applied to base stations and terminal devices. Figure 22 shows the training process. The inference process corresponding to Figure 22 can be found in Figure 17B or 18B. In Figure 22, the terminal device uses an independent model-based approach to obtain the first neural network model and the second neural network model (i.e., the fourth approach in Figure 7 and the independent model-based approach in Figure 5).

[0458] Figure 23 is applied to the base station and the server. Figure 23 shows the training process. The inference process corresponding to Figure 23 can be found in Figure 15B or 16B. In Figure 23, the base station uses an independent model-based approach to obtain the first neural network model, and the server uses an independent model-based approach to obtain the second neural network model (i.e., the fifth approach in Figure 7 and the independent model-based approach in Figure 5).

[0459] Figure 24 is applied to base stations and terminal devices. Figure 24 shows the training process. The inference process corresponding to Figure 24 can be found in Figure 17B or 18B. In Figure 24, the base station uses an independent model-based approach to obtain the first neural network model, and the terminal device uses an independent model-based approach to obtain the second neural network model (i.e., the fifth approach in Figure 7 and the independent model-based approach in Figure 5).

[0460] The following explanations are provided in conjunction with the accompanying drawings.

[0461] Figure 15A is a flowchart of a neural network model training method provided in an embodiment of this application. This method can be applied to base stations and servers.

[0462] As shown in Figure 15A, the method includes S1501-S1506.

[0463] In S1501, the base station sends the first signaling to the server.

[0464] According to embodiments of this application, the first signaling can be used to instruct the server to send fifth and sixth information to the base station via a first method. Optionally, the first signaling can also be used to indicate the size of the fifth and sixth information.

[0465] As one implementation, the first approach may include at least one of the following: a GTP-U header or a first protocol between the base station and the server. The first protocol may include fields for indicating fifth information and sixth information.

[0466] In S1502, upon receiving the first signaling, the server obtains the fifth and sixth information.

[0467] In S1503, the server sends the fifth and sixth information to the base station via the first method.

[0468] In S1504, the base station uses the fifth and sixth information to train the third neural network model, thereby obtaining the first neural network model and the second neural network model.

[0469] In S1505, the base station sends the model parameters of the second neural network model to the server and sends the second signaling via a third method.

[0470] According to embodiments of this application, the second signaling can be used to indicate the interaction information between the base station and the server and the second mode of carrying the interaction information. Optionally, the second signaling can be used to indicate the second mode, the interaction information, and the size of the interaction information.

[0471] As one implementation, the second approach may include at least one of the following: a GTP-U header or a third protocol between the base station and the server. The third protocol may include fields for indicating the second information.

[0472] As one implementation, the third approach may include at least one of the following: a GTP-U header or a fifth protocol between the base station and the server. The fifth protocol may include fields for indicating the second signaling.

[0473] In step 1506, the server deploys the second neural network model based on the model parameters of the second neural network model.

[0474] Based on Figure 15A, Figure 15B is a flowchart of a method for obtaining first information provided in an embodiment of this application. This method can be applied to base stations and servers.

[0475] As shown in Figure 15B, the method may include S1507-S1510.

[0476] In S1507, the server retrieves the fourth piece of information.

[0477] In S1508, the server inputs the fourth information into the second neural network model to obtain the second information.

[0478] In S1509, the server sends second information to the base station via a second method.

[0479] In S1510, the base station obtains the first information based on the first neural network model and the meaning of the second information and the interaction information.

[0480] Based on Figure 15A, Figure 15C is a flowchart illustrating another method for obtaining first information provided in an embodiment of this application. This method can be applied to base stations and servers.

[0481] As shown in Figure 15C, the method may include S1511-S1513.

[0482] In S1511, the server obtains the fourth piece of information.

[0483] In S1512, the server sends the fourth message to the base station.

[0484] In S1513, the base station inputs the fourth information into the third neural network model to obtain the first information.

[0485] Figure 16A is a flowchart of another neural network model training method provided in an embodiment of this application. This method can be applied to base stations and servers.

[0486] As shown in Figure 16A, the method may include S1601-S1606.

[0487] In S1601, the server retrieves the fifth and sixth pieces of information.

[0488] In S1602, the server uses the fifth and sixth information to train the third neural network model, thereby obtaining the first neural network model and the second neural network model.

[0489] In S1603, the server determines the interaction information between the base station and the server, and the second method for carrying the interaction information.

[0490] In S1604, the server sends the model parameters of the first neural network model to the base station and sends interactive information through a second method.

[0491] In S1605, the base station deploys the first neural network model based on the model parameters of the first neural network model.

[0492] In S1606, the base station determines the meaning of the interactive information.

[0493] It should be noted that the execution order of S1602 and S1603 is not limited in this embodiment. For example, S1602 can be executed first, then S1603 can be executed first, then S1602 can be executed second, and S1602 and S1603 can be executed simultaneously. Similarly, the execution order of S1605 and S1606 is not limited in this embodiment. For example, S1605 can be executed first, then S1606 can be executed third, then S1605 can be executed first, and S1605 and S1606 can be executed simultaneously.

[0494] Based on Figure 16A, Figure 16B is a flowchart of another method for obtaining first information provided in an embodiment of this application. This method can be applied to base stations and servers.

[0495] As shown in Figure 16B, the method may include S1607-S1610.

[0496] In S1607, the server retrieves the fourth piece of information.

[0497] In S1608, the server inputs the fourth information into the second neural network model to obtain the second information.

[0498] In S1609, the server sends second information to the base station via a second method.

[0499] In S1610, the base station obtains the first information based on the first neural network model and the meaning of the second information and the interaction information.

[0500] Based on Figure 16A, Figure 16C is a flowchart of another method for obtaining first information provided in an embodiment of this application. This method can be applied to base stations and servers.

[0501] As shown in Figure 16C, the method may include S1611-S1613.

[0502] In S1611, the server obtains the fourth piece of information.

[0503] In S1612, the server inputs the fourth information into the third neural network model to obtain the first information.

[0504] In S1613, the server sends the first information to the base station.

[0505] By establishing a source map between the base station and the server, the base station can transmit data packets according to a pre-determined transmission method based on the anticipated size of the first data packet, thus improving the reliability of air interface transmission and enhancing user experience. In this scenario, it increases the number of users a base station can serve, even with limited air interface resources, thereby improving air interface resource utilization.

[0506] Figure 17A is a flowchart of another neural network model training method provided in an embodiment of this application. This method can be applied to base stations and terminal devices.

[0507] As shown in Figure 17A, the method includes S1701-S1706.

[0508] In S1701, the base station sends the first signaling to the terminal device.

[0509] According to embodiments of this application, the first signaling can be used to instruct the terminal device to send fifth and sixth information to the base station via a first method. Optionally, the first signaling can also be used to indicate the size of the fifth and sixth information. Optionally, the first signaling can also be used to indicate the transmission period. The transmission period can be used to indicate the period during which the terminal device sends the fifth and sixth information to the base station. The transmission period can be configured using RRC or triggered via MAC CE. The trigger can include a single trigger or a periodic trigger after a single trigger.

[0510] As one implementation, the first approach may include at least one of the following: a second protocol between a MAC CE or a base station and a terminal device. The second protocol may include fields for indicating fifth information and sixth information.

[0511] In S1702, upon receiving the first signaling, the terminal device acquires the fifth and sixth information.

[0512] In S1703, the terminal device sends the fifth and sixth information to the base station via the first method.

[0513] In S1704, the base station uses the fifth and sixth information to train the third neural network model, thereby obtaining the first neural network model and the second neural network model.

[0514] In S1705, the base station sends the model parameters of the second neural network model to the terminal device and sends the second signaling through a third method.

[0515] According to embodiments of this application, the second signaling can be used to indicate the interaction information between the base station and the terminal device and the second mode of carrying the interaction information. Optionally, the second signaling can be used to indicate the second mode, the interaction information, and the size of the interaction information.

[0516] As one implementation, the second approach may include at least one of the following: UCI, UAI, or a fourth protocol between the base station and the terminal device. The fourth protocol may include fields for indicating the second information.

[0517] As one implementation, the third approach may include at least one of the following: MAC CE, DCI, or a sixth protocol between the base station and the terminal equipment. The sixth protocol may include fields for indicating the second signaling.

[0518] In 1706, the terminal device deploys the second neural network model based on the model parameters of the second neural network model.

[0519] Based on Figure 17A, Figure 17B is a flowchart of another method for obtaining first information provided by an embodiment of this application. This method can be applied to base stations and terminal devices.

[0520] As shown in Figure 17B, the method may include S1707-S1710.

[0521] In S1707, the terminal device obtains the fourth piece of information.

[0522] In S1708, the terminal device inputs the fourth information into the second neural network model to obtain the second information.

[0523] In S1709, the terminal device sends second information to the base station via a second method.

[0524] In S1710, the base station obtains the first information based on the first neural network model and the meaning of the second information and the interaction information.

[0525] Based on Figure 17A, Figure 17C is a flowchart of another method for obtaining first information provided by an embodiment of this application. This method can be applied to base stations and terminal devices.

[0526] As shown in Figure 17C, the method may include S1711-S1713.

[0527] In S1711, the terminal device obtains the fourth piece of information.

[0528] In S1712, the terminal device sends the fourth information to the base station.

[0529] In S1713, the base station inputs the fourth information into the third neural network model to obtain the first information.

[0530] Figure 18A is a flowchart of another neural network model training method provided in an embodiment of this application. This method can be applied to base stations and terminal devices.

[0531] As shown in Figure 18A, the method may include S1801-S1806.

[0532] In S1801, the terminal device obtains the fifth and sixth information.

[0533] In S1802, the terminal device uses the fifth and sixth information to train the third neural network model, thereby obtaining the first neural network model and the second neural network model.

[0534] In S1803, the terminal device determines the interaction information between the base station and the terminal device and the second method for carrying the interaction information.

[0535] In S1804, the terminal device sends the model parameters of the first neural network model to the base station and sends interactive information through a second method.

[0536] In S1805, the base station deploys the first neural network model based on the model parameters of the first neural network model.

[0537] In S1806, the base station determines the meaning of the interactive information.

[0538] Based on Figure 18A, Figure 18B is a flowchart of another method for obtaining first information provided in an embodiment of this application. This method can be applied to base stations and terminal devices.

[0539] As shown in Figure 18B, the method may include S1807-S1810.

[0540] In S1807, the terminal device obtains the fourth piece of information.

[0541] In S1808, the terminal device inputs the fourth information into the second neural network model to obtain the second information.

[0542] In S1809, the terminal device sends second information to the base station via a second method.

[0543] In S1810, the base station obtains the first information based on the first neural network model and the meaning of the second information and the interaction information.

[0544] Based on Figure 18A, Figure 18C is a flowchart of another method for obtaining first information provided by an embodiment of this application. This method can be applied to base stations and terminal devices.

[0545] As shown in Figure 18C, the method may include S1811-S1813.

[0546] In S1811, the terminal device obtains the fourth piece of information.

[0547] In S1812, the terminal device inputs the fourth information into the third neural network model to obtain the first information.

[0548] In S1813, the terminal device sends the first information to the base station.

[0549] By establishing a source map between the base station and the terminal device, the base station can transmit data packets according to a pre-determined transmission method based on the anticipated size of the first data packet, thus improving the reliability of air interface transmission and enhancing user experience. In this scenario, it increases the number of users a base station can serve, even with limited air interface resources, thereby improving air interface resource utilization.

[0550] Figure 19 is a flowchart of another neural network model training method provided in an embodiment of this application. This method can be applied to base stations and servers.

[0551] As shown in Figure 19, the method includes S1901-S1906.

[0552] In S1901, the base station sends the first signaling to the server.

[0553] According to embodiments of this application, the first signaling can be used to instruct the server to send fifth, eighth, and sixth information to the base station via a first method. Optionally, the first signaling can also be used to indicate the size of the fifth, eighth, and sixth information.

[0554] As one implementation method, the first method may include at least one of the following: a GTP-U packet header or a first protocol between the base station and the server. The first protocol may include fields for indicating the fifth information, the eighth information, and the sixth information.

[0555] In S1902, upon receiving the first signaling, the server obtains the fifth, eighth, and sixth information.

[0556] In S1903, the server sends the fifth, eighth, and sixth information to the base station via the first method.

[0557] In S1904, the base station uses the fifth and eighth information to train the second neural network model to obtain the trained second neural network model, and uses the eighth and sixth information to train the first neural network model to obtain the trained first neural network model.

[0558] In S1905, the base station sends the model parameters of the second neural network model to the server and sends the second signaling via a third method.

[0559] According to embodiments of this application, the second signaling can be used to indicate the interaction information between the base station and the server and the second mode of carrying the interaction information. Optionally, the second signaling can be used to indicate the second mode, the interaction information, and the size of the interaction information.

[0560] As one implementation, the second approach may include at least one of the following: a GTP-U header or a third protocol between the base station and the server. The third protocol may include fields for indicating the second information.

[0561] As one implementation, the third approach may include at least one of the following: a GTP-U header or a fifth protocol between the base station and the server. The fifth protocol may include fields for indicating the second signaling.

[0562] In 1906, the server deployed the second neural network model based on the model parameters of the second neural network model.

[0563] Based on Figure 19, the first information can be obtained using the methods shown in Figures 15B or 16B, which will not be elaborated here.

[0564] By establishing a source map between the base station and the server, the base station can transmit data packets according to a pre-determined transmission method based on the anticipated size of the first data packet, thus improving the reliability of air interface transmission and enhancing user experience. In this scenario, it increases the number of users a base station can serve, even with limited air interface resources, thereby improving air interface resource utilization.

[0565] Figure 20 is a flowchart of another neural network model training method provided in an embodiment of this application. This method can be applied to base stations and servers.

[0566] As shown in Figure 20, the method may include S2001-S2006.

[0567] In S2001, the server retrieves the fifth, eighth, and sixth pieces of information.

[0568] In S2002, the server uses the fifth and eighth information to train the second neural network model, obtaining the trained second neural network model, and uses the eighth and sixth information to train the first neural network model, obtaining the trained first neural network model.

[0569] In S2003, the server determines the interaction information between the base station and the server, and the second method for carrying the interaction information.

[0570] In S2004, the server sends the model parameters of the first neural network model to the base station and sends interactive information through a second method.

[0571] In S2005, the base station deploys the first neural network model based on the model parameters of the first neural network model.

[0572] In S2006, the base station determines the meaning of the interactive information.

[0573] Based on Figure 20, the first information can be obtained using the methods shown in Figures 15B or 15C, which will not be elaborated here.

[0574] By establishing a source map between the base station and the server, the base station can transmit data packets according to a pre-determined transmission method based on the anticipated size of the first data packet, thus improving the reliability of air interface transmission and enhancing user experience. In this scenario, it increases the number of users a base station can serve, even with limited air interface resources, thereby improving air interface resource utilization.

[0575] Figure 21 is a flowchart of another neural network model training method provided in an embodiment of this application. This method can be applied to base stations and terminal devices.

[0576] As shown in Figure 21, the method includes S2101-S2106.

[0577] In S2101, the base station sends the first signaling to the terminal device.

[0578] According to embodiments of this application, the first signaling can be used to instruct the terminal device to send fifth, eighth, and sixth information to the base station via a first method. Optionally, the first signaling can also be used to indicate the size of the fifth, eighth, and sixth information. Optionally, the first signaling can also be used to indicate the transmission period. The transmission period can be used to indicate the period during which the terminal device sends the fifth, eighth, and sixth information to the base station. The transmission period can be configured using RRC or triggered via MAC CE. The trigger can include a single trigger or a periodic trigger after a single trigger.

[0579] As one implementation, the first approach may include at least one of the following: a second protocol between a MAC CE or a base station and a terminal device. The second protocol may include fields for indicating the fifth information, the eighth information, and the sixth information.

[0580] In S2102, upon receiving the first signaling, the terminal device acquires the fifth, eighth, and sixth information.

[0581] In S2103, the terminal device sends the fifth, eighth, and sixth information to the base station via the first method.

[0582] In S2104, the base station uses the fifth and eighth information to train the second neural network model to obtain the trained second neural network model, and uses the eighth and sixth information to train the first neural network model to obtain the trained first neural network model.

[0583] In S2105, the base station sends the model parameters of the second neural network model to the terminal device and sends the second signaling through a third method.

[0584] According to embodiments of this application, the second signaling can be used to indicate the interaction information between the base station and the terminal device and the second mode of carrying the interaction information. Optionally, the second signaling can be used to indicate the second mode, the interaction information, and the size of the interaction information.

[0585] As one implementation, the second approach may include at least one of the following: UCI, UAI, or a fourth protocol between the base station and the terminal device. The fourth protocol may include fields for indicating the second information.

[0586] As one implementation, the third approach may include at least one of the following: MAC CE, DCI, or a sixth protocol between the base station and the terminal equipment. The sixth protocol may include fields for indicating the second signaling.

[0587] In 2106, the terminal device deploys the second neural network model based on the model parameters of the second neural network model.

[0588] Based on Figure 21, the first information can be obtained using the methods shown in Figures 17B or 18B, which will not be elaborated here.

[0589] Figure 22 is a flowchart of another neural network model training method provided in an embodiment of this application. This method can be applied to base stations and terminal devices.

[0590] As shown in Figure 22, the method may include S2201-S2206.

[0591] In S2201, the terminal device acquires the fifth, eighth, and sixth information.

[0592] In S2202, the terminal device uses the fifth and eighth information to train the second neural network model to obtain the trained second neural network model, and uses the eighth and sixth information to train the first neural network model to obtain the trained first neural network model.

[0593] In S2203, the terminal device determines the interaction information between the base station and the terminal device and the second method for carrying the interaction information.

[0594] In S2204, the terminal device sends the model parameters of the first neural network model to the base station and sends interactive information through a second method.

[0595] In S2205, the base station deploys the first neural network model based on the model parameters of the first neural network model.

[0596] In S2206, the base station determines the meaning of the interactive information.

[0597] Based on Figure 22, the first information can be obtained using the methods shown in Figure 17B or 18B, which will not be elaborated here.

[0598] By establishing a source map between the base station and the terminal device, the base station can transmit data packets according to a pre-determined transmission method based on the anticipated size of the first data packet, thus improving the reliability of air interface transmission and enhancing user experience. In this scenario, it increases the number of users a base station can serve, even with limited air interface resources, thereby improving air interface resource utilization.

[0599] Figure 23 is a flowchart of another neural network model training method provided in an embodiment of this application. This method can be applied to base stations and servers.

[0600] As shown in Figure 23, the method includes S2301-S2306.

[0601] In S2301, the base station sends the first signaling to the server.

[0602] According to embodiments of this application, the first signaling can be used to instruct the server to send eighth and sixth information to the base station via a first method. Optionally, the first signaling can also be used to indicate the size of the eighth and sixth information.

[0603] As one implementation, the first approach may include at least one of the following: a GTP-U header or a first protocol between the base station and the server. The first protocol may include fields for indicating the eighth information and fields for indicating the sixth information.

[0604] In S2302, upon receiving the first signaling, the server obtains the eighth and sixth information.

[0605] In S2303, the server uses the fifth and sixth information to train the second neural network model, thus obtaining the trained second neural network model.

[0606] In S2304, the server sends the eighth and sixth information to the base station via the first method.

[0607] In S2305, the base station uses the eighth and sixth information to train the first neural network model, thus obtaining the trained first neural network model.

[0608] In S2306, the base station sends the second signaling via a third method.

[0609] According to embodiments of this application, the second signaling can be used to indicate the interaction information between the base station and the server and the second mode of carrying the interaction information. Optionally, the second signaling can be used to indicate the second mode, the interaction information, and the size of the interaction information.

[0610] As one implementation, the second approach may include at least one of the following: a GTP-U header or a third protocol between the base station and the server. The third protocol may include fields for indicating the second information.

[0611] As one implementation, the third approach may include at least one of the following: a GTP-U header or a fifth protocol between the base station and the server. The fifth protocol may include fields for indicating the second signaling.

[0612] Based on Figure 23, the first information can be obtained using the methods shown in Figures 15B or 16B, which will not be elaborated here.

[0613] By establishing a source map between the base station and the server, the base station can transmit data packets according to a pre-determined transmission method based on the anticipated size of the first data packet, thus improving the reliability of air interface transmission and enhancing user experience. In this scenario, it increases the number of users a base station can serve, even with limited air interface resources, thereby improving air interface resource utilization.

[0614] Figure 24 is a flowchart of another neural network model training method provided in an embodiment of this application. This method can be applied to base stations and terminal devices.

[0615] As shown in Figure 24, the method includes S2401-S2406.

[0616] In S2401, the base station sends the first signaling to the terminal device.

[0617] According to embodiments of this application, the first signaling can be used to instruct the terminal device to send eighth information and sixth information to the base station via a first method. Optionally, the first signaling can also be used to indicate the size of the eighth information and the size of the sixth information.

[0618] As one implementation, the first approach may include at least one of the following: a second protocol between a MAC CE or a base station and a terminal device. The second protocol may include fields for indicating the eighth information and fields for indicating the sixth information.

[0619] In S2402, upon receiving the first signaling, the terminal device acquires the eighth and sixth information.

[0620] In S2403, the terminal device uses the fifth and sixth information to train the second neural network model, thus obtaining the trained second neural network model.

[0621] In S2404, the terminal device sends the eighth and sixth information to the base station via the first method.

[0622] In S2405, the base station uses the eighth and sixth information to train the first neural network model, thus obtaining the trained first neural network model.

[0623] In S2406, the base station sends the second signaling via a third method.

[0624] According to embodiments of this application, the second signaling can be used to indicate the interaction information between the base station and the terminal device and the second mode of carrying the interaction information. Optionally, the second signaling can be used to indicate the second mode, the interaction information, and the size of the interaction information.

[0625] As one implementation, the second approach may include at least one of the following: UCI, UAI, or a fourth protocol between the base station and the terminal device. The fourth protocol may include fields for indicating the second information.

[0626] As one implementation, the third approach may include at least one of the following: MAC CE, DCI, or a sixth protocol between the base station and the terminal equipment. The sixth protocol may include fields for indicating the second signaling.

[0627] Based on Figure 24, the first information can be obtained using the methods shown in Figures 17B or 18B, which will not be elaborated here.

[0628] By establishing a source map between the base station and the terminal device, the base station can transmit data packets according to a pre-determined transmission method based on the anticipated size of the first data packet, thus improving the reliability of air interface transmission and enhancing user experience. In this scenario, it increases the number of users a base station can serve, even with limited air interface resources, thereby improving air interface resource utilization.

[0629] The methods provided in the embodiments of this application have been described in detail above with reference to several accompanying drawings. The apparatus provided in the embodiments of this application will now be described with reference to the accompanying drawings.

[0630] Based on the same concept as the aforementioned embodiments of the data transmission method applied to the second device side, this application also provides a first data transmission device 2500, which can be deployed on the second device side to implement the data transmission method applied to the second device side provided in this application. The first data transmission device 2500 includes units or modules for implementing the various steps of the data transmission method.

[0631] Figure 25 is a schematic block diagram of a first data transmission device provided in an embodiment of this application.

[0632] As shown in Figure 25, the first data transmission device 2500 may include an acquisition module 2510 and a determination module 2520.

[0633] The acquisition module 2510 is used to acquire first information at a first moment. The first information is used to indicate relevant information of the first data packet.

[0634] The determining module 2520 is used to determine the first transmission mode of the first data packet based on the first information at a second time. The second time is not earlier than the first time. The first transmission mode is used to transmit the first data packet according to the first transmission mode when the first data packet is obtained at a third time, and the third time is later than the second time.

[0635] The first data transmission device 2500 according to the embodiments of this application can be used to execute the data transmission method applied to the second device side as described in the embodiments of this application. The above and other operations and / or functions of each module in the first data transmission device 2500 correspond to the data transmission method of the embodiments of this application. For the sake of brevity, they will not be described again here.

[0636] Based on the same concept as the aforementioned embodiments of the data transmission method applied to the first device side, this application also provides a second data transmission device 2600, which can be deployed on the first device side to implement the data transmission method applied to the first device side provided in this application. The second data transmission device 2600 includes units or modules for implementing various operations in the data transmission method.

[0637] Figure 26 is a schematic block diagram of a second data transmission device provided in an embodiment of this application.

[0638] As shown in Figure 26, the second data transmission device 2600 may include a transmission module 2610.

[0639] The sending module 2610 is used to send second information to the second device side. The second information is used by the second device side to acquire first information at a first moment. The first information is determined by the second device side based on a first neural network model, according to the second information and / or third information. At a second moment, the second device side determines a first transmission method for the first data packet based on the first information. The second moment is no earlier than the first moment. The first transmission method is used to transmit the first data packet according to the first transmission method if the first data packet is acquired at a third moment. The third moment is later than the second moment.

[0640] The sending module 2610 is used to send fourth information to the second device side. The fourth information is used by the second device side to obtain first information at a first moment (the first information is determined by the second device side based on the fourth information), and at a second moment to determine a first transmission method for the first data packet based on the first information. This first transmission method is used to transmit the first data packet if it is obtained at a third moment. Alternatively...

[0641] The sending module 2610 is used to send first information to the second device side. The first information is used by the second device side to obtain the first information at a first moment, and at a second moment to determine the first transmission mode of the first data packet based on the first information;

[0642] According to embodiments of this application, first information is used to indicate relevant information of a first data packet. Second information is used to indicate information related to the first information. Third information is used to indicate at least one of the following: first channel information or first available resources. First channel information is used to indicate information related to a first channel. The first channel is used to transmit the first data packet. The first available resources are used to indicate available resources on the second device side. Fourth information is used to indicate information related to the first information.

[0643] It is understood that the division of modules or units in the above-described device is merely a logical functional division. Each function can correspond to a functional module or unit, or two or more functions can be integrated into one functional module or unit. In actual implementation, all or some units or modules can be integrated into a single physical entity, or they can be distributed across different physical entities. Furthermore, the aforementioned functional modules or units can be implemented in hardware, software, or a combination of both. Whether a function is executed in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0644] Figure 27 is a schematic block diagram of an electronic device provided in an embodiment of this application.

[0645] As shown in Figure 27, the electronic device 2700 includes one or more processors 2710. The processor 2710 may be a general-purpose processor or a special-purpose processor, etc.

[0646] Optionally, if the electronic device 2700 can be a terminal device or a RAN node, one or more processors 2710 may include a baseband processor, also known as a modem processor.

[0647] Optionally, when the electronic device 2700 can be a terminal device or a RAN node, the electronic device 2700 may include a radio frequency (RF) processing system and at least one antenna. In the downlink or sidelink direction, the RF processing system receives RF signals through the antenna and transmits the RF-processed signals to one or more processors 2710 for further processing. In the uplink or sidelink direction, the processor 2710 may transmit terminal-side information processed by one or more processors 2410 to the RF processing system. The RF processing system transmits the RF-processed signals through the antenna.

[0648] In one example, the radio frequency (RF) processing system, serving as the communication interface for external communication of terminal devices or RAN nodes, may include an RF front end (RFFE) and an RF transceiver (RFT). The RFFE can be used to perform at least several processing operations, such as shaping, passband selection, or gain, on RF signals received by the antenna or RF signals to be transmitted through the antenna. The RFFE may include at least one of the following components: an RF switch, a duplexer, a filter, a power amplifier, an antenna tuner, or a low-noise amplifier. The RFFE can be a circuit system composed of multiple discrete devices or integrated into one or more chips. The RF transceiver is used to process the RF signals received by the RFFE into baseband / IF signals for further processing by one or more processors 2710, and to process a baseband / IF signal provided by one or more processors 2710 into an RF signal for transmission to the RFFE. The baseband / IF signals transmitted between the RF transceiver and one or more processors 2710 can be digital or analog signals. The RF transceiver can be implemented by one or more chips, typically referred to as an RF integrated circuit (RFIC).

[0649] Optionally, the RF transceiver and RF front-end can be packaged in a single chip. In one example, the RF transceiver, RF front-end, and baseband processor can also be packaged in a single chip.

[0650] Optionally, if the electronic device 2700 can be a terminal device, it may further include at least one of a voice system, a multimedia system, or an interface circuit. The voice system can be used to process audio signals. The multimedia system can be used to handle multimedia-related operations, such as video encoding / decoding or image processing. The interface circuit can be used to enable communication with other terminal components. For example, other terminal components may include at least one of a display, an input device, or a memory.

[0651] Optionally, if the electronic device 2700 can be a terminal, one or more processors 2710 may include an application processor for processing the terminal operating system and application layer.

[0652] Optionally, the baseband processor may include one or more processor cores and interface circuitry. The one or more processor cores may be used to process signals and execute one or more communication protocols. Optionally, the baseband processor may also include memory. The memory may be used to store at least a portion of the corresponding computer program instructions and / or data. In one example, one or more processor cores implement the relevant steps in the above method embodiments by executing the computer program instructions stored in the memory. In this application embodiment, the memory may be used to store the corresponding computer program instructions and / or data. This can mean that the memory is used to store all the corresponding computer program instructions and / or data for the processor core to execute, or it can mean that the memory is used to store a portion of the corresponding computer program instructions and / or data. This portion of the corresponding computer program instructions and / or data may include the computer program instructions and / or data that the processor core currently needs to execute. The memory can store different portions of computer program instructions and / or data multiple times for the processor core to execute in order to implement the relevant steps in the above method embodiments. The interface circuit serves as a communication interface for communication with other components, such as transmitting signals with the RF processing system, communicating with the application processor, voice system, and / or multimedia system via a bus, and / or communicating with related components of the voice system and / or multimedia system via a bus, for example, transmitting data control signals with the application processor. Optionally, to reduce the load on the processor core, the baseband processor may also include baseband signal processing circuitry to perform at least some baseband signal processing tasks, such as demodulation, modulation, encoding, or decoding. Optionally, one or more processors 2710, the voice system, the multimedia system, and the interface circuitry may be packaged into a single processor chip. For example, a SoC (System on Clip) chip or a SIP (System in Package) chip. In one example, the above may also be packaged into multiple chips. For example, the baseband processor may be packaged as a single chip, or packaged with some or all of the circuitry of the RF processing system into a single chip.

[0653] Optionally, the memory can be on-chip memory, for example, located on the processor chip.

[0654] Optionally, the memory can be off-chip memory, for example, located outside the processor chip.

[0655] Optionally, in one example, processor 2710 may include a computer program (also referred to as code or instructions) that can be run on processor 2710, causing electronic device 2700 to perform the methods performed on the second device side or the first device side in the method embodiments described above. In yet another possible design, electronic device 2700 includes circuitry (not shown in FIG27) for implementing the functions of the second device side or the first device side in the method embodiments described above.

[0656] Optionally, the electronic device 2700 may include one or more memories 2720 storing computer programs (sometimes referred to as code or instructions) that can be run on the processor 2710, causing the electronic device 2700 to perform the methods performed on the second device side or the first device side in the above embodiments.

[0657] Optionally, the processor 2710 and / or memory 2720 may also store data. The processor 2710 and memory 2720 may be configured separately or integrated together. Optionally, the electronic device 2700 may also include a communication interface 2730. The processor 2710, sometimes referred to as a processing unit, controls the electronic device 2700 (e.g., on the second device side or the first device side). The communication interface 2730, sometimes referred to as a transceiver unit, transceiver, transceiver circuit, transceiver, or input / output interface, is used to implement the transceiver function of the electronic device 2700; for example, the communication interface 2730 can be used to receive first information, second information, or fourth information.

[0658] Optionally, the processor 2710 and the communication interface 2730 are coupled to each other.

[0659] When the electronic device 2700 is used to implement the device embodiment described above, the processor 2710 can be used to execute the function of the determination module 2520, and the communication interface 2730 can be used to execute the function of the acquisition module 2510. Whether the communication interface 2730 is used for sending or receiving depends on whether the electronic device 2700 is used to perform a sending or receiving action in the scheme it executes.

[0660] It is understood that when the electronic device 2700 is a second device side or a terminal side, the communication interface 2730 can be a transceiver, specifically including a transmitter and a receiver, with the transmitter used to send signals and the receiver used to receive signals.

[0661] Optionally, the electronic device 2700 also includes a power supply circuit that can be used to power the electronic device 2700.

[0662] The above-described method embodiments can be applied to a processor, or implemented by a processor. A processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method embodiments can be completed through integrated logic circuits in the processor's hardware or through software instructions.

[0663] The aforementioned processors, application processors, baseband processors, processor circuits, or processor cores can be collectively referred to as processors. These processors may include Central Processing Units (CPUs), Microprocessor Units (MPUs), Microcontroller Units (MCUs), Graphics Processing Units (GPUs), Artificial Intelligence Processors (AIPs), Neural Processing Units (NPUs), Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or any combination thereof.

[0664] The steps of the method disclosed in the embodiments of this application can be directly manifested as being executed by a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in mature storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0665] The memory in the embodiments of this application may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), phase-change memory (PCM), resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), hard disk drive (HDD), or flash memory. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), and synchronous dynamic random access memory (DRAM). DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Synchronous Link DRAM (SLDRAM), and Direct Rambus RAM (DR RAM). Furthermore, volatile memory may also include registers and / or caches. It should be noted that the memory used in the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0666] In one example, the computer program instructions for performing the above embodiments may be stored in non-volatile memory, such as at least a portion of memory 2720 (e.g., at least one of ROM, flash memory, EPROM, or hard disk). When electronic device 2700 is running, the corresponding computer program instructions may be partially or entirely loaded into memory with a faster transfer speed than processor 2710 (e.g., at least one of RAM, SRAM, DRAM, PCM, ReRAM, MRAM, FRAM, cache, or registers) for processor execution to implement the steps in the above method embodiments.

[0667] This application also provides a chip system including at least one processor for supporting the implementation of the functions of the second device side or the first device side involved in any of the above method embodiments.

[0668] In one possible design, the chip system also includes a memory for storing computer program instructions and data, which may be located inside or outside the processor.

[0669] The chip system can consist of chips or include chips and other discrete components.

[0670] This application also provides a computer program product, which includes a computer program (also referred to as code or instructions), wherein when the computer program is run, the method executed on the second device side or the method executed on the first device side in the embodiments of this application is executed.

[0671] This application also provides a computer-readable storage medium storing a computer program (also referred to as code or instructions). When the computer program is executed, the method executed on the second device side or the method executed on the first device side in the embodiments of this application is executed.

[0672] This application also provides a data transmission system, which includes the aforementioned first data transmission device and second data transmission device.

[0673] The methods provided in the above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any combination thereof. When implemented in software, they can be implemented, in whole or in part, as a computer program product. This computer program product may include one or more computer instructions. When these computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic disks), optical media (e.g., DVDs), or semiconductor media (e.g., solid state disks (SSDs)).

[0674] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0675] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0676] In the several embodiments provided in this application, it should be understood that the disclosed systems, electronic devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0677] The unit described as a separate component may or may not be physically separate. The component shown as a unit may or may not be a physical unit; that is, it may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0678] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0679] If this function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, or part of it, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks.

[0680] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more blocks of the flowchart illustrations and / or one or more blocks of the block diagrams. In the various embodiments of this application, unless otherwise specified or in accordance with logical conflicts, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0681] Obviously, those skilled in the art can make various modifications and variations to the embodiments of this application without departing from the scope of this application. Therefore, if these modifications and variations to the embodiments of this application fall within the scope of the claims of this application and their equivalents, the embodiments of this application are also intended to include these modifications and variations.

Claims

1. A data transmission method, characterized in that, include: At a first moment, first information is acquired, wherein the first information is used to indicate relevant information of the first data packet; At a second moment, a first transmission method for the first data packet is determined based on the first information, wherein the second moment is not earlier than the first moment, and the first transmission method is used to transmit the first data packet according to the first transmission method when the first data packet is obtained at a third moment, wherein the third moment is later than the second moment.

2. The method according to claim 1, characterized in that, The acquisition of the first information includes: Based on a first neural network model, the first information is determined according to second information and / or third information, wherein the second information is related to the first information, and the third information is used to indicate at least one of the following: first channel information or first available resources, wherein the first channel information is used to indicate information related to a first channel, the first channel is used to transmit the first data packet, and the first available resources are used to indicate available resources on the second device side; or Receive fourth information from the first device side, and determine the first information based on the fourth information, wherein the fourth information is related to the first information; or Receive the first information from the first device side.

3. The method according to claim 2, characterized in that, The second information is determined by the first device based on the fourth information, which indicates at least one of the following: a first complexity or a first instruction, wherein the first complexity indicates the complexity of the first screen, and the first instruction indicates the instruction that triggers the first screen; or The first information is determined by the first device based on the fourth information.

4. The method according to claim 3, characterized in that, When the first data packet is a video data packet or an image data packet, the first complexity is used to indicate at least one of the following: first image entropy, first image variance, first edge density, first gradient magnitude, first image contrast, first color contrast, first color histogram, first color entropy, or first fractal dimension; or When the first data packet is an audio data packet, the first complexity is used to indicate at least one of the following: first audio entropy, first spectral entropy, first dynamic range, or first audio variance.

5. The method according to any one of claims 1-4, characterized in that, The second information is used to indicate at least one of the following: the predicted size corresponding to the first data packet, a first prediction confidence level, a first prediction confidence interval, a prediction priority, a predicted arrival time, a second prediction confidence level, a second prediction confidence interval, a predicted service type, or a predicted quality of service requirement, wherein the first prediction confidence level indicates the confidence level of the predicted size, the first prediction confidence interval indicates the confidence interval of the predicted size, the second prediction confidence level indicates the confidence level of the predicted arrival time, and the second prediction confidence interval indicates the confidence interval of the predicted arrival time; and / or The first information is used to indicate at least one of the following: the predicted size corresponding to the first data packet, the first prediction confidence level, the first prediction confidence interval, the prediction priority, the predicted arrival time, the second prediction confidence level, the second prediction confidence interval, the predicted service type, or the predicted service quality requirement; The first information and the second information are represented in different ways.

6. The method according to any one of claims 2-5, characterized in that, The first neural network model was obtained in the following way: The first neural network model is the first part of the third neural network model, which is obtained in the following way: The third neural network model is trained by the second device side or the first device side using the fifth information and the sixth information. The fifth information is used to indicate at least one of the following: a second complexity or a second instruction. The second complexity is used to indicate the complexity of the second screen. The second instruction is used to indicate the instruction that triggers the second screen. The sixth information is used to indicate the actual information related to the second data packet. or The third neural network model is trained by the second device side or the first device side using the fifth information, the sixth information, and the seventh information, wherein the seventh information is used to indicate at least one of the following: second channel information or second available resources; the second channel information is used to indicate information related to the second channel; the second channel is used to transmit the second data packet; and the second available resources are used to indicate available resources on the second device side or available resources on other second device sides; or The first neural network model is trained by the second device side or the first device side using the eighth information and the sixth information, wherein the eighth information is used to indicate information related to the sixth information; or The first neural network model is trained by the second device side or the first device side using the eighth information, the seventh information and the sixth information.

7. The method according to claim 6, characterized in that, When the second data packet is a video data packet or an image data packet, the second complexity is used to indicate at least one of the following: second image entropy, second image variance, second edge density, second gradient magnitude, second image contrast, second color contrast, second color histogram, second color entropy, or second fractal dimension; when the second data packet is an audio data packet, the second complexity is used to indicate at least one of the following: second audio entropy, second spectral entropy, second dynamic range, or second audio variance; and / or The sixth information is used to indicate at least one of the following: the actual size corresponding to the second data packet, the first actual confidence level, the first actual confidence interval, the actual priority, the actual arrival time, the second actual confidence level, the second actual confidence interval, the actual service type, or the actual service quality requirement. The first actual confidence level is used to indicate the confidence level of the actual size, the first actual confidence interval is used to indicate the confidence interval of the actual size, the second actual confidence level is used to indicate the confidence level of the actual arrival time, and the second actual confidence interval is used to indicate the confidence interval of the actual arrival time. and / or The eighth piece of information is used to indicate at least one of the following: the actual size corresponding to the second data packet, the first actual confidence level, the first actual confidence interval, the actual priority, the actual arrival time, the second actual confidence level, the second actual confidence interval, the actual service type, or the actual quality of service requirement; The sixth and eighth pieces of information are represented in different ways.

8. The method according to claim 6 or 7, characterized in that, When the first neural network model is trained by the second device, the second device sends a first signaling to the first device. The first signaling is used to instruct the first device to send the fifth information and the sixth information to the second device in a first manner.

9. The method according to claim 8, characterized in that, When the first device side is the service side and the second device side is the access side, the first method includes at least one of the following: a GPRS tunneling protocol GTP-U header for the user plane or a first protocol between the access side and the service side, wherein the first protocol includes a field for indicating the fifth information and a field for indicating the sixth information; or When the first device side is the terminal side and the second device side is the access side, the first method includes at least one of the following: a Media Intervention Control Layer Control Unit (MAC CE) or a second protocol between the access side and the terminal side, wherein the second protocol includes fields for indicating the fifth information and fields for indicating the sixth information.

10. The method according to claim 8 or 9, characterized in that, The first signaling is also used to indicate the size of the fifth information and the size of the sixth information; and / or When the first device side is the terminal side and the second device side is the access side, the first signaling is also used to indicate the transmission period, which is used to indicate the period during which the terminal side sends the fifth information and the sixth information to the access side.

11. The method according to any one of claims 2-10, characterized in that, The second information is received by the second device from the first device via a second method; Where the first device side is the service side and the second device side is the access side, the second method includes at least one of the following: the GTP-U packet header or a third protocol between the access side and the service side, wherein the third protocol includes a field for indicating the second information; or When the first device side is the terminal side and the second device side is the access side, the second method includes at least one of the following: uplink control information (UCI), user assistance information (UAI), or a fourth protocol between the access side and the terminal side, wherein the fourth protocol includes a field for indicating the second information.

12. The method according to claim 11, characterized in that, When the first neural network model is trained by the second device side, the second device side sends a second signaling to the first device side. The second signaling is used to indicate the second mode and the interaction information between the second device side and the first device side, or the second signaling is used to indicate the second mode, the interaction information and the size of the interaction information, where the second information is the interaction information.

13. The method according to claim 12, characterized in that, The second device sends the second signaling to the first device via a third method; Where the first device side is the service side and the second device side is the access side, the third method includes at least one of the following: the GTP-U packet header or the fifth protocol between the access side and the service side, the fifth protocol including a field for indicating the second signaling; or When the first device side is the terminal side and the second device side is the access side, the third method includes at least one of the following: MAC CE, downlink control information DCI, or a sixth protocol between the access side and the terminal side, wherein the sixth protocol includes a field for indicating the second signaling.

14. The method according to any one of claims 1-13, characterized in that, The first transmission method is used to transmit the first data packet according to the second transmission method when the first data packet is acquired at the third time, wherein the second transmission method is determined according to the first transmission method and the first data packet.

15. An information transmission method, comprising: Send second information to the second device side, wherein the second information is used by the second device side to acquire first information at a first time, the first information being determined by the second device side based on a first neural network model according to the second information and / or third information; at a second time, determine a first transmission method for the first data packet based on the first information, the second time being no earlier than the first time; the first transmission method is used to transmit the first data packet according to the first transmission method if the first data packet is acquired at a third time, the third time being later than the second time; or Send fourth information to the second device side, wherein the fourth information is used by the second device side to obtain the first information at the first time, the first information being determined by the second device side based on the fourth information, and at the second time to determine the first transmission method of the first data packet based on the first information, the first transmission method being used to transmit the first data packet according to the first transmission method when the first data packet is obtained at the third time; or Send first information to the second device side, wherein the first information is used by the second device side to obtain the first information at the first moment, and at the second moment to determine the first transmission mode of the first data packet based on the first information; Wherein, the first information is used to indicate relevant information of the first data packet, the second information is used to indicate information related to the first information, the third information is used to indicate at least one of the following: first channel information or first available resources, the first channel information is used to indicate information related to the first channel, the first channel is used to transmit the first data packet, the first available resources are used to indicate available resources on the second device side, and the fourth information is related to the first information.

16. The method according to claim 15, characterized in that, The second information is determined based on the fourth information, which indicates at least one of the following: a first complexity or a first instruction, wherein the first complexity indicates the complexity of the first screen, and the first instruction indicates the instruction that triggers the first screen; or The first information is determined based on the fourth information.

17. The method according to claim 16, characterized in that, When the first data packet is an audio data packet, the first complexity is used to indicate at least one of the following: first image entropy, first image variance, first edge density, first gradient magnitude, first image contrast, first color contrast, first color histogram, first color entropy, or first fractal dimension; or When the first data packet is an audio data packet, the first complexity is used to indicate at least one of the following: first audio entropy, first spectral entropy, first dynamic range, or first audio variance.

18. The method according to any one of claims 15-17, characterized in that, The second information is used to indicate at least one of the following: the predicted size corresponding to the first data packet, a first prediction confidence level, a first prediction confidence interval, a prediction priority, a predicted arrival time, a second prediction confidence level, a second prediction confidence interval, a predicted service type, or a predicted quality of service requirement, wherein the first prediction confidence level indicates the confidence level of the predicted size, the second prediction confidence interval indicates the confidence interval of the predicted size, the second prediction confidence level indicates the confidence level of the predicted arrival time, and the second prediction confidence interval indicates the confidence interval of the predicted arrival time; and / or The first information is used to indicate at least one of the following: the predicted size corresponding to the first data packet, the first prediction confidence level, the first prediction confidence interval, the prediction priority, the predicted arrival time, the second prediction confidence level, the second prediction confidence interval, the predicted service type, or the predicted service quality requirement; The first information and the second information are represented in different ways.

19. The method according to any one of claims 16-18, characterized in that, The second information is determined based on the fourth information, including: The second information is obtained by inputting the fourth information into the third neural network model, wherein the third neural network model is obtained in the following manner: The third neural network model is the second part of the second neural network model, which was obtained in the following way: The third neural network model is trained on the second device side or the first device side using fifth and sixth information. The fifth information is used to indicate at least one of the following: a second complexity or a second instruction, wherein the second complexity indicates the complexity of the second screen, the second instruction indicates the instruction that triggers the second screen, and the sixth information indicates actual information related to the second data packet; or The third neural network model is trained by the second device side or the first device side using the fifth information, the sixth information, and the seventh information, wherein the seventh information is used to indicate at least one of the following: second channel information or second available resources; the second channel information is used to indicate information related to the second channel; the second channel is used to transmit the second data packet; and the second available resources are used to indicate available resources on the second device side or available resources on other second device sides; or The second neural network model is trained on the second device side or the first device side using the fifth and eighth information, wherein the eighth information is used to indicate information related to the sixth information; or The second neural network model is trained by the second device side or the first device side using the fifth information, the seventh information, and the eighth information.

20. The method according to claim 19, characterized in that, When the second data packet is a video data packet or an image data packet, the second complexity is used to indicate at least one of the following: second image entropy, second image variance, second edge density, second gradient magnitude, second image contrast, second color contrast, second color histogram, second color entropy, or second fractal dimension; when the second data packet is an audio data packet, the second complexity is used to indicate at least one of the following: second audio entropy, second spectral entropy, second dynamic range, or second audio variance; and / or The sixth piece of information is used to indicate at least one of the following: the actual size corresponding to the second data packet, the first actual confidence level, the first actual confidence interval, the actual priority, the actual arrival time, the second actual confidence level, the second actual confidence interval, the actual service type, or the actual service quality requirement. The first actual confidence level is used to indicate the confidence level of the actual size, the first actual confidence interval is used to indicate that the actual size is false, the second actual confidence level is used to indicate the confidence level of the actual arrival time, and the second actual confidence interval is used to indicate the confidence interval of the actual arrival time. and / or The eighth piece of information is used to indicate at least one of the following: the actual size corresponding to the second data packet, the first actual confidence level, the first actual confidence interval, the actual priority, the actual arrival time, the second actual confidence level, the second actual confidence interval, the actual service type, or the actual quality of service requirement; The sixth and eighth pieces of information are represented in different ways.

21. The method according to claim 19 or 20, characterized in that, When the second neural network model is trained by the second device side, the first device side receives a first signaling sent by the second device side, wherein the first signaling is used to instruct the first device side to send the fifth information and the sixth information to the second device side in a first manner.

22. The method according to claim 21, characterized in that, When the first device side is the service side and the second device side is the access side, the first method includes at least one of the following: a GPRS tunneling protocol GTP-U header for the user plane or a first protocol between the access side and the service side, wherein the first protocol includes a field for indicating the fifth information and a field for indicating the sixth information; or When the first device side is the terminal side and the second device side is the access side, the first method includes at least one of the following: a Media Intervention Control Layer Control Unit (MAC CE) or a second protocol between the access side and the terminal side, wherein the second protocol includes fields for indicating the fifth information and fields for indicating the sixth information.

23. The method according to claim 21 or 22, characterized in that, The first signaling is also used to indicate the size of the fifth information and the size of the sixth information; and / or When the first device side is the terminal side and the second device side is the access side, the first signaling is also used to indicate the transmission period, which is used to indicate the period during which the terminal side sends the fifth information and the sixth information to the access side.

24. The method according to any one of claims 15-23, characterized in that, Sending the second information to the second device includes: The second information is sent to the second device via a second method; Where the first device side is the service side and the second device side is the access side, the second method includes at least one of the following: the GTP-U packet header or a third protocol between the access side and the service side, wherein the third protocol includes a field for indicating the second information; or When the first device side is the terminal side and the second device side is the access side, the second method includes at least one of the following: uplink control information (UCI), user assistance information (UAI), or a fourth protocol between the access side and the terminal side, wherein the fourth protocol includes a field for indicating the second information.

25. The method according to claim 24, characterized in that, When the second neural network model is trained by the second device side, the first device side receives a second signaling from the second device side. The second signaling is used to indicate the second mode and the interaction information between the second device side and the first device side, or the second signaling is used to indicate the second mode, the interaction information and the size of the interaction information, where the second information is the interaction information.

26. An electronic device, characterized in that, The processor includes a processor coupled to a memory for storing computer programs, and the processor for executing the computer programs stored in the memory. So that the electronic device performs the method as described in any one of claims 1-14; or, So that the electronic device performs the method as described in any one of claims 15-25.

27. A computer-readable storage medium, characterized in that, The computer stores instructions that, when executed on the computer, cause the computer to perform the method as described in any one of claims 1-14, or cause the computer to perform the method as described in any one of claims 15-25.

28. A computer program product, characterized in that, The computer program product includes: a computer program that, when run, causes a computer to perform the method of any one of claims 1-14, or causes a computer to perform the method of any one of claims 15-25.

29. A chip system, characterized in that, The chip system is applied to an electronic device, the chip system including one or more processors, the one or more processors being configured to invoke computer instructions to cause the electronic device to perform the method as described in any one of claims 1-14, or to cause the electronic device to perform the method as described in any one of claims 15-25.

30. A data transmission system, characterized in that, Including the first equipment side and the second equipment side, The second device side is used to perform the method as described in any one of claims 1-14, and the first device side is used to perform the method as described in any one of claims 15-25.

Images