Communication method, terminal, device, system, medium, and computer program product

Abstract

The present disclosure relates to a communication method, a terminal, a device, a system, a medium, and a computer program product. The communication method executed by a terminal comprises: receiving a first received signal, the first received signal being a signal received by the terminal after a first pilot signal sent by a network device is transmitted via a channel; inputting the first received signal into a trained first partial model to obtain first information outputted by the first partial model; and sending the first information to the network device, the first information being used by the network device as an input of a trained second partial model to obtain first channel state information outputted by the second partial model. In this way, an error of channel state information obtained by a network device can be reduced, thereby improving communication reliability and efficiency.

Description

Communication method, terminal, device, system, medium and computer program product TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of communication, and particularly relates to a communication method, a terminal, a device, a system, a medium and a computer program product. BACKGROUND

[0002] With the development of Artificial Intelligence (AI) technology and Machine Learning (ML) technology, in the field of communication technology, AI / ML technology can be used to assist communication between terminals and network devices.

[0003] SUMMARY

[0004] The present disclosure provides a communication method, a terminal, a device, a system, a medium and a computer program product.

[0005] In a first aspect, a communication method is provided, which is performed by a terminal, and includes: receiving a first received signal, the first received signal being a signal received by the terminal after a first pilot signal transmitted by a network device is transmitted through a channel; inputting the first received signal into a first part model which is trained, to obtain first information output by the first part model; and sending the first information to the network device, the first information being used by the network device as an input of a second part model which is trained, to obtain first channel state information output by the second part model.

[0006] In a second aspect, a communication method is provided, which is performed by a network device, and includes: receiving first information sent by a terminal, the first information being obtained by the terminal after a first received signal received by the terminal is input into a first part model which is trained, wherein the first received signal is a signal received by the terminal after a first pilot signal transmitted by the network device is transmitted through a channel; and inputting the first information into a second part model which is trained, to obtain first channel state information output by the second part model.

[0007] In a third aspect, the embodiments of the present disclosure provide a terminal, comprising: a transceiver configured to receive a first received signal, the first received signal being a signal received by the terminal after a first pilot signal transmitted by a network device is transmitted through a channel; a processor configured to input the first received signal into a first part model trained, to obtain first information output by the first part model; and the transceiver is further configured to transmit the first information to the network device, the first information being used by the network device as an input of a second part model trained, to obtain first channel state information output by the second part model.

[0008] In a fourth aspect, the embodiments of the present disclosure provide a network device, comprising: a transceiver configured to receive first information transmitted by a terminal, the first information being obtained by the terminal after a first received signal received by the terminal is input into a first part model trained, wherein the first received signal is a signal received by the terminal after a first pilot signal transmitted by the network device is transmitted through a channel; and a processor configured to input the first information into a second part model trained, to obtain first channel state information output by the second part model.

[0009] In a fifth aspect, the embodiments of the present disclosure provide a communication device, comprising: one or more processors; and a memory coupled to the processors, the memory having stored thereon executable instructions that, when executed by the processors, cause the communication method of the first aspect or the second aspect to be performed.

[0010] In a sixth aspect, the embodiments of the present disclosure provide a communication system, comprising a terminal and a network device, wherein the terminal is configured to implement the communication method of the first aspect, and the network device is configured to implement the communication method of the second aspect.

[0011] In a seventh aspect, the embodiments of the present disclosure provide a storage medium, the storage medium storing instructions, when the instructions are executed on a communication device, causing the communication device to perform the communication method of the first aspect or the second aspect.

[0012] In an eighth aspect, the embodiments of the present disclosure provide a computer program product, comprising a computer program and / or instructions, when the computer program and / or instructions are executed by a communication device, implementing the communication method of the first aspect or the second aspect.

[0013] By adopting the technical solutions of the present disclosure, at least the following beneficial technical effects can be achieved:

[0014] The terminal inputs the received first received signal into the trained first part model to obtain first information output by the first part model. The terminal sends the first information to the network device, and the first information is used by the network device as an input of a trained second part model to obtain first channel state information output by the second part model. This way of jointly using the first part model of the terminal and the second part model of the network device realizes CSI feedback, can reduce the error of the CSI obtained by the network device, and the CSI with higher accuracy can make the communication system more suitable for the current channel condition, providing guarantee for high reliability and high rate communication. BRIEF DESCRIPTION OF DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the following describes the drawings required for the embodiments. The following drawings are only some embodiments of the present disclosure, and do not specifically limit the protection scope of the present disclosure.

[0016] FIG. 1A is one exemplary schematic diagram of an architecture of a communication system according to an embodiment of the present disclosure.

[0017] FIG. 1B is a schematic diagram of a channel state information feedback process according to an embodiment of the present disclosure.

[0018] FIG. 1C is a schematic diagram of a model architecture according to an embodiment of the present disclosure.

[0019] FIG. 2A is an interaction schematic diagram of a communication method according to an embodiment of the present disclosure.

[0020] FIG. 2B is a functional schematic diagram of a first part model according to an embodiment of the present disclosure.

[0021] FIG. 2C is a functional schematic diagram of a first part model according to an embodiment of the present disclosure.

[0022] FIG. 2D is a schematic diagram of training a bilateral model according to an embodiment of the present disclosure.

[0023] FIG. 2E is a schematic diagram of training a bilateral model according to an embodiment of the present disclosure.

[0024] FIG. 3A is a flow schematic diagram of a communication method according to an embodiment of the present disclosure.

[0025] FIG. 3B is a flow schematic diagram of model training according to an embodiment of the present disclosure.

[0026] FIG. 4 is a flow schematic diagram of a communication method according to an embodiment of the present disclosure.

[0027] FIG. 5 is a schematic diagram of sending a pilot signal according to an embodiment of the present disclosure.

[0028] FIG. 6 is a schematic diagram illustrating transmitting a pilot signal according to an embodiment of the present disclosure.

[0029] FIG. 7A is a schematic diagram illustrating a structure of a terminal according to an embodiment of the present disclosure.

[0030] FIG. 7B is a schematic diagram illustrating a structure of a network device according to an embodiment of the present disclosure.

[0031] FIG. 8A is a schematic diagram illustrating a structure of a communication device according to an embodiment of the present disclosure.

[0032] FIG. 8B is a schematic diagram illustrating a structure of a chip according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0033] Embodiments of the present disclosure provide a communication method, a terminal, a device, a system, a medium and a computer program product.

[0034] In a first aspect, embodiments of the present disclosure provide a communication method performed by a terminal, the method comprising: receiving a first received signal, the first received signal being a signal received by the terminal after a first pilot signal transmitted by a network device passes through a channel; inputting the first received signal into a first part model trained, to obtain first information output by the first part model; and transmitting the first information to the network device, the first information being used by the network device as an input of a second part model trained, to obtain first channel state information output by the second part model.

[0035] In the above embodiment, the terminal inputs the received first received signal into the first part model trained, to obtain the first information output by the first part model. The terminal transmits the first information to the network device, and the first information is used by the network device as an input of the second part model trained, to obtain the first channel state information output by the second part model. This way of combining the first part model of the terminal and the second part model of the network device realizes CSI feedback, which can reduce the error of the CSI obtained by the network device, and the more accurate CSI can make the communication system more suitable for the current channel condition, providing a guarantee for high reliability and high rate communication.

[0036] In some embodiments of the first aspect, the first part model and the second part model are components of a bilateral model.

[0037] In the above embodiment, the bilateral model including the first part model of the terminal and the second part model of the network device realizes CSI compressed feedback by combining multiple processes in channel estimation, channel feedback, channel coding and data modulation, which can reduce the influence of channel estimation error and improve the accuracy of the channel state information obtained by the network device.

[0038] In some embodiments of the first aspect, in some embodiments, the first channel state information comprises first full channel information and / or a first eigenvector corresponding to the first full channel information.

[0039] In the above embodiments, it is specified that the first channel state information can comprise first full channel information and / or a first eigenvector.

[0040] In some embodiments of the first aspect, in some embodiments, the first information is at least one of the following:

[0041] N bits, N being an integer greater than or equal to 1;

[0042] M complex numbers, M being an integer greater than or equal to 1;

[0043] K real numbers, K being an integer greater than or equal to 1.

[0044] In the above embodiments, the data form of the first information of the first part model output is specified.

[0045] In some embodiments of the first aspect, in some embodiments, the double-sided model is obtained by training the first part model to be trained, a channel model, and the second part model to be trained in collaboration according to a sample data set, to obtain the first part model trained and the second part model trained; wherein the channel model is used to simulate a wireless channel through which the first information passes.

[0046] In the above embodiments, by training the first part model to be trained, the channel model, and the second part model to be trained in collaboration, a double-sided model with high robustness can be obtained. Among them, because the first part model and the second part model are trained using the channel model simulating the wireless channel. This enables them to learn effective CSI estimation, CSI compression and CSI reconstruction strategies, while taking into account the effects of channel attenuation and noise. By incorporating the channel model into the training process, the first part model and the second part model trained can adapt to real wireless channel environments, making the double-sided model more robust.

[0047] In some embodiments of the first aspect, in some embodiments, the sample data set comprises input sample data and output sample data corresponding to the input sample data; and the collaborative training of the first partial model to be trained, the channel model, and the second partial model to be trained according to the sample data set to obtain the trained first partial model and the trained second partial model comprises: performing at least one first process until a trained model convergence condition is met to obtain the trained first partial model and the trained second partial model; the first process comprises: inputting the input sample data into the first partial model; inputting output data of the first partial model into the channel model; inputting output data of the channel model into the second partial model; and adjusting a model parameter of at least one of the first partial model, the channel model, and the second partial model according to an error between output data of the second partial model and the output sample data.

[0048] In the above embodiments, the training method can be used to obtain a bilateral model with good robustness and high accuracy.

[0049] In some embodiments of the first aspect, in some embodiments, the trained model convergence condition comprises at least one of the following:

[0050] The number of times of performing the first process is greater than or equal to a first threshold value;

[0051] A loss function error of the trained model is less than or equal to a second threshold value.

[0052] In the above embodiments, the trained model convergence condition can avoid overfitting of the bilateral model on the sample data set.

[0053] In some embodiments of the first aspect, in some embodiments, the sample data set comprises measured sample data and / or synthetic sample data.

[0054] In the above embodiments, training the bilateral model with a rich sample data set can improve the robustness of the model.

[0055] In some embodiments of the first aspect, in some embodiments, the sample data set comprises at least one of the following:

[0056] Second received signals and corresponding second channel state information, wherein the second received signals are signals received by the terminal in a first frequency domain range;

[0057] a third receive signal and corresponding third channel state information, wherein the third receive signal is a signal received by the terminal after the network device transmits a signal from an antenna port in a first antenna port range and the signal passes through a channel.

[0058] In the above embodiment, since the sample data set includes data of frequency domain correlation and / or spatial domain correlation, training the model according to the sample data set can enable the bilateral model to learn the correlation features of the corresponding domain, thereby improving the performance of the bilateral model in the frequency domain and / or the spatial domain.

[0059] In some embodiments of the first aspect, in some embodiments, the second channel state information includes second full channel information corresponding to a second frequency domain range and / or a second feature vector corresponding to the second full channel information, wherein the second frequency domain range at least includes the first frequency domain range.

[0060] In the above embodiment, since the second receive signal in the first frequency domain range corresponds to the second channel state information of the second frequency domain range, and the second frequency domain range is greater than or equal to the first frequency domain range, training the bilateral model according to such a sample data set can enable the bilateral model to learn the ability to obtain channel state information corresponding to a larger frequency domain range from a receive signal in a smaller frequency domain range, thereby improving the performance of the bilateral model in the frequency domain.

[0061] In some embodiments of the first aspect, in some embodiments, the third channel state information includes third full channel information corresponding to a second antenna port range and / or a third feature vector corresponding to the third full channel state information, wherein the second antenna port range at least includes the first antenna port range.

[0062] In the above embodiment, since the third receive signal in the first antenna port range corresponds to the third channel state information of the second antenna port range, and the second antenna port range is greater than or equal to the first antenna port range, training the bilateral model according to such a sample data set can enable the bilateral model to learn the ability to obtain channel state information corresponding to a larger spatial domain range from a receive signal in a smaller spatial domain range, thereby improving the performance of the bilateral model in the spatial domain.

[0063] In some embodiments of the first aspect, in some embodiments, the interval between two adjacent frequency domain positions for receiving the second receive signal is pre-set or indicated by a first device, wherein the first device includes the terminal and / or the network device.

[0064] In some embodiments of the first aspect, in some embodiments, the interval between two adjacent spatial positions for receiving the third received signal is pre-configured or indicated by the first device, the first device comprising the terminal and / or the network device.

[0065] In some embodiments of the first aspect, in some embodiments, the number of the sample data sets is a plurality, and the plurality of sample data sets correspond to a plurality of continuous time points.

[0066] In the above embodiments, the bilateral model is trained by the plurality of sample data sets of the plurality of continuous time points, so that the bilateral model learns the features of the time domain correlation, and the bilateral model can be applied to a channel estimation scenario based on the time domain correlation, thereby improving the robustness of the bilateral model.

[0067] In a second aspect, the embodiments of the present disclosure provide a communication method, performed by a network device, the method comprising: receiving first information sent by a terminal, the first information being obtained by inputting a first received signal received by the terminal into a trained first part model, wherein the first received signal is a signal received by the terminal after a first pilot signal sent by the network device is transmitted through a channel; inputting the first information into a trained second part model to obtain first channel state information output by the second part model.

[0068] In some embodiments of the second aspect, in some embodiments, the first part model and the second part model are components of a bilateral model.

[0069] In some embodiments of the second aspect, in some embodiments, the first channel state information comprises first full channel information and / or a first feature vector corresponding to the first full channel information.

[0070] In some embodiments of the second aspect, in some embodiments, the first information is at least one of the following:

[0071] N bits, N being an integer greater than or equal to 1;

[0072] M complex numbers, M being an integer greater than or equal to 1;

[0073] K real numbers, K being an integer greater than or equal to 1.

[0074] In some embodiments of the second aspect, the bilateral model is trained by performing collaborative training on the first partial model to be trained, a channel model, and the second partial model to be trained according to a sample data set, to obtain the trained first partial model and the trained second partial model; the channel model is used to simulate a wireless channel through which the first information passes.

[0075] In some embodiments of the second aspect, the sample data set includes input sample data and output sample data corresponding to the input sample data.

[0076] The collaborative training on the first partial model to be trained, the channel model, and the second partial model to be trained according to a sample data set, to obtain the trained first partial model and the trained second partial model, includes performing at least one first process until a trained model convergence condition is met, to obtain the trained first partial model and the trained second partial model; the first process includes inputting the input sample data into the first partial model; inputting output data of the first partial model into the channel model; inputting output data of the channel model into the second partial model; and adjusting a model parameter of at least one of the first partial model, the channel model, and the second partial model according to an error between output data of the second partial model and the output sample data.

[0077] In some embodiments of the second aspect, the trained model convergence condition includes at least one of the following:

[0078] The number of times of performing the first process is greater than or equal to a first threshold value.

[0079] A loss function error of the trained model is less than or equal to a second threshold value.

[0080] In some embodiments of the second aspect, the sample data set includes measured sample data and / or synthetic sample data.

[0081] In some embodiments of the second aspect, the sample data set includes at least one of the following:

[0082] Second received signals and corresponding second channel state information, wherein the second received signals are signals received by the terminal in a first frequency domain range;

[0083] a third receive signal and corresponding third channel state information, wherein the third receive signal is a signal received by the terminal after the network device transmits a signal from an antenna port in a first antenna port range and the signal is transmitted through a channel.

[0084] In some embodiments combined with the second aspect, in some embodiments, the second channel state information comprises second full channel information corresponding to a second frequency domain range and / or a second eigenvector corresponding to the second full channel information, wherein the second frequency domain range at least comprises the first frequency domain range.

[0085] In some embodiments combined with the second aspect, in some embodiments, the third channel state information comprises third full channel information corresponding to a second antenna port range and / or a third eigenvector corresponding to the third full channel information, wherein the second antenna port range at least comprises the first antenna port range.

[0086] In some embodiments combined with the second aspect, in some embodiments, a spacing between two adjacent frequency domain positions for receiving the second receive signal is pre-configured or indicated by a first device, wherein the first device comprises the terminal and / or the network device.

[0087] In some embodiments combined with the second aspect, in some embodiments, a spacing between two adjacent spatial domain positions for receiving the third receive signal is pre-configured or indicated by a first device, wherein the first device comprises the terminal and / or the network device.

[0088] In some embodiments combined with the second aspect, in some embodiments, the number of the sample data sets is a plurality, and the plurality of sample data sets correspond to a plurality of continuous time points.

[0089] In a third aspect, the embodiments of the present disclosure provide a terminal, comprising at least one of a transceiver module and a processing module; wherein the terminal is configured to perform the optional implementation manners of the first aspect.

[0090] In a fourth aspect, the embodiments of the present disclosure provide a network device, comprising at least one of a transceiver module and a processing module; wherein the network device is configured to perform the optional implementation manners of the second aspect.

[0091] In a fifth aspect, the embodiments of the present disclosure provide a communication device, comprising: one or more processors; wherein the communication device is configured to perform the optional implementation manners of the first aspect and / or the second aspect.

[0092] In a sixth aspect, the embodiments of the present disclosure provide a communication system, comprising: a terminal, a network device; wherein the terminal is configured to perform the method described in the optional implementation manner of the first aspect, and the network device is configured to perform the method described in the optional implementation manner of the second aspect.

[0093] In a seventh aspect, the embodiments of the present disclosure provide a storage medium, which stores instructions, when the instructions run on a communication device, cause the communication device to perform the method described in the optional implementation manner of the first aspect and / or the second aspect.

[0094] In an eighth aspect, the embodiments of the present disclosure provide a program product, which is executed by a communication device, causes the communication device to perform the method described in the optional implementation manner of the first aspect and / or the second aspect.

[0095] In a ninth aspect, the embodiments of the present disclosure provide a computer program, when it runs on a computer, causes the computer to perform the method described in the optional implementation manner of the first aspect and / or the second aspect.

[0096] In a tenth aspect, the embodiments of the present disclosure provide a chip or chip system. The chip or chip system comprises processing circuitry configured to perform the method described in the optional implementation manner of the first aspect and / or the second aspect.

[0097] It can be understood that the terminal, network device, communication device, communication system, storage medium, program product, computer program, chip or chip system are all used to perform the method proposed by the embodiments of the present disclosure. Therefore, the beneficial effects they can achieve can refer to the beneficial effects in the corresponding method, which will not be repeated here.

[0098] The embodiments of the present disclosure provide a communication method, terminal, device, system, medium and computer program product. In some embodiments, the terms of the communication method, information processing method, AI model based joint channel estimation and CSI feedback method can be replaced with each other; the terms of the communication device, information processing device, AI model based joint channel estimation and CSI feedback device can be replaced with each other; the terms of the communication system and information processing system, AI model based joint channel estimation and CSI feedback system can be replaced with each other.

[0099] The embodiments of the present disclosure are not exhaustive, but only illustrate some embodiments, and are not specific limitations on the protection scope of the present disclosure. In the case of no contradiction, each step in an embodiment can be implemented as an independent embodiment, and the steps can be combined arbitrarily, for example, the scheme after removing part of the steps in an embodiment can also be implemented as an independent embodiment, and the order of the steps in an embodiment can be exchanged arbitrarily, in addition, the optional implementation manners in an embodiment can be combined arbitrarily; in addition, the embodiments can be combined arbitrarily, for example, part or all steps of different embodiments can be combined arbitrarily, an embodiment can be combined with optional implementation manners of other embodiments arbitrarily.

[0100] In each embodiment of the present disclosure, the terms and / or descriptions between the embodiments are consistent if there is no special description and logical conflict, and can be referred to each other, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0101] The terms used in the embodiments of the present disclosure are only for the purpose of describing the specific embodiments, and not as a limitation on the present disclosure.

[0102] In the embodiments of the present disclosure, unless otherwise specified, the elements expressed in singular form, such as "one", "a", "the", "above", "said", "preceding", "this" and the like, can represent "one and only one", and can also represent "one or more", "at least one" and the like. For example, in the case of using articles such as "a", "an", "the" and the like in English, the noun after the article can be understood as singular expression, and can also be understood as plural expression.

[0103] In the embodiments of the present disclosure, "plurality" means two or more.

[0104] In some embodiments, the terms "at least one of", "one or more", "a plurality of", "multiple" and the like can be replaced with each other.

[0105] In some embodiments, the description of "at least one of A, B", "A and / or B", "in a case A, in another case B", "in response to a case A, in response to a case B", and the like, can include the following technical solutions according to the case: in some embodiments, A (A is executed regardless of B); in some embodiments, B (B is executed regardless of A); in some embodiments, A and B are selectively executed (A and B are selected from A and B); in some embodiments, A and B (A and B are executed). When there are more branches such as A, B, C, and the like, the above is similar.

[0106] In some embodiments, the description of "A or B" and the like can include the following technical solutions according to the case: in some embodiments, A (A is executed regardless of B); in some embodiments, B (B is executed regardless of A); in some embodiments, A and B are selectively executed (A and B are selected from A and B). When there are more branches such as A, B, C, and the like, the above is similar.

[0107] The prefix words "first", "second", and the like in the embodiments of the present disclosure are only used to distinguish different description objects, and do not constitute a limitation on the position, order, priority, quantity, or content of the description objects. The description of the description objects should refer to the description in the context of the claims or embodiments, and should not be limited by the prefix words. For example, the description object is "field", and the ordinal words before "field" in "first field" and "second field" do not limit the position or order between "fields". "First" and "second" do not limit whether the "fields" they modify are in the same message, nor do they limit the order of "first field" and "second field". For another example, the description object is "level", and the ordinal words before "level" in "first level" and "second level" do not limit the priority between "levels". For another example, the quantity of the description object is not limited by the ordinal words, and can be one or more. For example, "first device", where the quantity of "device" can be one or more. In addition, the objects modified by different prefix words can be the same or different, for example, the description object is "device", and "first device" and "second device" can be the same device or different devices, and their types can be the same or different; for another example, the description object is "information", and "first information" and "second information" can be the same information or different information, and their contents can be the same or different.

[0108] In some embodiments, "including A", "containing A", "for indicating A", "carrying A" can be interpreted as directly carrying A, or indirectly indicating A.

[0109] In some embodiments, the terms "time / frequency", "time / frequency domain", and the like refer to the time domain and / or the frequency domain.

[0110] In some embodiments, the terms "in response to", "in response to determining", "in the case of", "when", "if", "if", etc. can be replaced with each other.

[0111] In some embodiments, the terms "greater than", "greater than or equal to", "not less than", "more than", "more than or equal to", "not less than", "higher than", "higher than or equal to", "not lower than", "above", etc. can be replaced with each other, and the terms "less than", "less than or equal to", "not greater than", "less than", "less than or equal to", "not more than", "lower than", "lower than or equal to", "not higher than", "below", etc. can be replaced with each other.

[0112] In some embodiments, the device, etc. can be interpreted as an entity, and can also be interpreted as virtual, and the name is not limited to the name described in the embodiments. The terms "device", "equipment", "device", "circuit", "network element", "node", "function", "unit", "section", "system", "network", "chip", "chip system", "entity", "subject", etc. can be replaced with each other.

[0113] In some embodiments, "network" can be interpreted as a device (e.g., access network device, core network device, etc.) contained in the network.

[0114] In some embodiments, the terms “access network device (AN device),” “radio access network device (RAN device),” “base station (BS),” “radio base station,” “fixed station,” “node,” “access point,” “transmission point (TP),” “reception point (RP),” “transmission / reception point (TRP),” “panel,” “antenna panel,” “antenna array,” “cell,” “macro cell,” “small cell,” “femto cell,” “pico cell,” “sector,” “cell group,” “serving cell,” “carrier,” “component carrier,” “bandwidth part (BWP),” and the like can be used interchangeably.

[0115] In some embodiments, the terms "terminal," "terminal device," "user equipment (UE)," "user terminal," "mobile station (MS)," "mobile terminal (MT)," "subscriber station," "mobile unit," "subscriber unit," "wireless unit," "remote unit," "mobile device," "wireless device," "wireless communication device," "remote device," "mobile subscriber station," "access terminal," "mobile terminal," "wireless terminal," "remote terminal," "handset," "user agent," "mobile client," "client," and so on can be replaced with each other.

[0116] In some embodiments, the access network device, the core network device, or the network device can be replaced with a terminal. For example, the embodiments of the present disclosure can also be applied to a structure in which communication between the access network device, the core network device, or the network device and the terminal is replaced with communication between a plurality of terminals (e.g., device-to-device (D2D), vehicle-to-everything (V2X), etc.). In this case, the terminal can also be configured to have all or part of the functions of the access network device. In addition, the terms "uplink," "downlink," and the like can also be replaced with terms corresponding to the inter-terminal communication (e.g., "side"). For example, the uplink channel, the downlink channel, and the like can be replaced with the side channel, and the uplink, the downlink, and the like can be replaced with the sidelink.

[0117] In some embodiments, the terminal can be replaced with the access network device, the core network device, or the network device. In this case, the access network device, the core network device, or the network device can also be configured to have all or part of the functions of the terminal.

[0118] In some embodiments, obtaining data, information, and the like can comply with laws and regulations of the country where the location is situated.

[0119] In some embodiments, data, information, and the like can be obtained after obtaining consent of a user.

[0120] In addition, each element, each row, or each column in the table of the embodiments of the present disclosure can be implemented as an independent embodiment, and any combination of any element, any row, or any column can also be implemented as an independent embodiment.

[0121] FIG. 1A is a schematic diagram of an architecture of a communication system according to an embodiment of the present disclosure. As shown in FIG. 1A, the communication system 100 can include a terminal 101 and a network device 102.

[0122] In some embodiments, the terminal 101 includes at least one of a mobile phone, a wearable device, an Internet of Things device, a car with communication function, a smart car, a Pad, a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in self-driving, a wireless terminal device in remote medical surgery, a wireless terminal device in smart grid, a wireless terminal device in transportation safety, a wireless terminal device in smart city, a wireless terminal device in smart home, and the like, but is not limited thereto.

[0123] In some embodiments, the network device 102 can include at least one of an access network device and a core network device.

[0124] Optionally, the network device 102 is an access network device. Optionally, the access network device is at least one of a node or device that accesses a terminal to a wireless network, and can include at least one of an evolved NodeB (eNB) in a 5G communication system, a next generation eNB (ng-eNB), a next generation NodeB (gNB), a node B (NB), a home node B (HNB), a home evolved node B (HeNB), a wireless backhaul device, a radio network controller (RNC), a base station controller (BSC), a base transceiver station (BTS), a base band unit (BBU), a mobile switching center, a base station in a 6G communication system, an Open RAN, a Cloud RAN, a base station in other communication systems, an access node in a Wi-Fi system, but is not limited thereto.

[0125] In some embodiments, the network device 102 is a base station. Optionally, the base station is at least one of a macro base station, a micro base station (also referred to as a small station), a relay station, an access point, a 5G base station or a future base station, a satellite, a Transmitting and Receiving Point (TRP), a Transmitting Point (TP), a mobile switching center, or other devices that perform a base station function in a communication system, etc., and the embodiments of the present disclosure are not limited thereto. For convenience of description, in all embodiments of the present disclosure, devices that provide a wireless communication function for a terminal device are collectively referred to as network devices or base stations.

[0126] In some embodiments, the network device 102 is a core network device. Optionally, the core network device can be one device including all or part of a first network element, a second network element, etc., or can be a plurality of devices or device groups including all or part of the first network element, the second network element, etc. The network element can be virtual or physical. The core network includes at least one of an Evolved Packet Core (EPC), a 5G Core Network (5GCN), a Next Generation Core (NGC), etc.

[0127] In some embodiments, the technical solutions of the present disclosure can be applied to an Open RAN architecture, at this time, the interfaces between or within the access network devices involved in the embodiments of the present disclosure can become internal interfaces of the Open RAN, and the processes and information interactions between these internal interfaces can be implemented through software or programs.

[0128] In some embodiments, the access network device can be composed of a central unit (CU) and a distributed unit (DU), where the CU can also be referred to as a control unit. The CU-DU structure can split the protocol layers of the access network device, and some of the functions of the protocol layers are controlled by the CU, and the remaining or all of the functions of the protocol layers are distributed in the DU and controlled by the CU. However, the present disclosure is not limited thereto.

[0129] It can be understood that the communication system described in the embodiments of the present disclosure is for more clearly illustrating the technical solutions of the embodiments of the present disclosure, and does not constitute a limitation on the technical solutions proposed by the embodiments of the present disclosure. It can be known by those skilled in the art that, as the system architecture evolves and new business scenarios appear, the technical solutions proposed by the embodiments of the present disclosure are also applicable to similar technical problems.

[0130] The following embodiments of the present disclosure can be applied to the communication system 100 shown in FIG. 1A or part of the subject, but are not limited thereto. The subjects shown in FIG. 1A are exemplary, and the communication system can include all or part of the subjects in FIG. 1A, or other subjects other than FIG. 1A. The number and form of each subject is arbitrary, each subject can be real or virtual, the connection relationship between each subject is exemplary, each subject can not be connected or can be connected, the connection can be in any way, can be direct connection or indirect connection, can be wired connection or wireless connection.

[0131] Embodiments of the present disclosure can be applied to Long Term Evolution (LTE), LTE-Advanced (LTE-A), LTE-Beyond (LTE-B), SUPER 3G, IMT-Advanced, 4th generation mobile communication system (4G), 5th generation mobile communication system (5G), 5G new radio (NR), Future Radio Access (FRA), New-Radio Access Technology (RAT), New Radio (NR), New radio access (NX), Future generation radio access (FX), Global System for Mobile communications (GSM (registered trademark)), CDMA2000, Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi (registered trademark)), IEEE 802.16 (WiMAX (registered trademark)), IEEE 802.20, Ultra-WideBand (UWB), Bluetooth (Bluetooth (registered trademark)), Public Land Mobile Network (PLMN) network, Device-to-Device (D2D) system, Machine to Machine (M2M) system, Internet of Things (IoT) system, Vehicle-to-Everything (V2X), system using other communication methods, next-generation system expanded based thereon, and the like. In addition, a plurality of systems can be combined (for example, combination of LTE or LTE-A and 5G, and the like).

[0132] In some embodiments, referring to FIG. 1B, for conventional CSI feedback, the terminal side at least includes a CSI compression encoding, channel encoding, modulation, and the like module, and then the UE reports the compressed CSI to the network through the uplink channel. The network side at least includes demodulation, channel decoding, and inverse operation to recover the CSI to obtain approximate pre-compression CSI.

[0133] In some embodiments, each of the above modules can implement the corresponding function through AI / ML model inference, which will result in more models deployed on the terminal side or network side, making the complexity of managing the models significantly increase. In order to alleviate this problem, the functions of multiple modules can be implemented through one AI / ML model.

[0134] In some embodiments, referring to FIG. 1C, the embodiments of the present disclosure propose a method of implementing joint CSI feedback and channel coding based on a bilateral AI / ML model. As can be seen from FIG. 1C, the bilateral model includes a CSI compression model and a CSI recovery model. The CSI compression model is implemented through an encoder, the input of the encoder is the CSI obtained by the UE through channel estimation, and the output information is obtained through a quantizer to obtain a binary bit stream. The bit stream passes through a channel network, which is used to simulate the characteristics of the wireless channel in the training process. For example, in the presence of additive white Gaussian noise (AWGN), a binary symmetric channel network can be deployed. The network introduces errors in the feedback bits according to the signal-to-noise ratio (SINR) of the channel. The output of the channel network is the input of the dequantizer, which recovers the compressed CSI data from the quantization level and converts it back to a floating-point value as the input of the CSI recovery model. The CSI recovery model is a decoder deployed on the network side, which aims to minimize the distortion introduced in the compression and quantization process to recover the original CSI information.

[0135] In some embodiments, in the training process of the bilateral model shown in FIG. 1C, the encoder and the decoder are trained using a simulated channel network. This enables them to learn effective compression and reconstruction strategies while taking into account the effects of channel attenuation and noise. By incorporating the channel network into the training process, the system becomes more robust to real wireless channel conditions.

[0136] In some embodiments, the bilateral model shown in FIG. 1C only considers the joint design of channel compression (referring to CSI compression) and channel coding, but the CSI input to the encoder is first estimated by the UE through the downlink pilot signal. Therefore, the performance of channel estimation will affect the performance of the joint design of channel compression and channel coding. In view of this, the embodiments of the present disclosure further propose a communication method, a terminal, a device, a system, a medium and a computer program product. The compression feedback of CSI is realized through a bilateral AI / ML model of joint channel estimation, channel feedback and channel coding to reduce the influence of channel estimation error.

[0137] FIG. 2A is an interaction diagram illustrating a communication method according to an embodiment of the present disclosure. As shown in FIG. 2A, the embodiments of the present disclosure relate to a communication method, which is performed by the communication system 100, and the above method comprises:

[0138] Step S201, the network device 102 sends a first pilot signal.

[0139] In some embodiments, the terminal receives a first received signal. The first received signal is a signal received by the terminal after the first pilot signal sent by the network device is transmitted through a channel. That is, the terminal receives the first pilot signal sent by the network device to obtain the first received signal.

[0140] It should be noted that in the embodiments of the present disclosure, the pilot signal sent by the sending end reaches the receiving end through a channel, and the signal received by the receiving end is referred to as a received signal.

[0141] In some embodiments, the first pilot signal is used for channel state information measurement, which can help the terminal to evaluate the channel quality and perform various measurement tasks of the downlink.

[0142] In some embodiments, the first pilot signal can be a channel state information reference signal (CSI-RS).

[0143] In some embodiments, the name of the first pilot signal is not limited, which is, for example, a reference signal, a measurement signal, a known signal, etc.

[0144] Step S202, the terminal 101 inputs the first received signal into the first part model to obtain the first information output by the first part model.

[0145] In some embodiments, the first part model is used to obtain the first information according to the first received signal.

[0146] In some embodiments, the first part model can be an AI model or an ML model, and the present disclosure does not make specific limitations thereon. Moreover, the name of the first part model is not limited, which is, for example, a CSI compression model, a first sub-model, an encoding model, an encoding network, etc.

[0147] In some embodiments, the first received signal is the input of the first part model. By inputting the first received signal into the first part model, the first information output by the first part model can be obtained, that is, the first information is the output of the first part model.

[0148] In some embodiments, the received first received signal can be directly input into the first part model.

[0149] In some embodiments, the first received signal can be pre-processed, and the pre-processed signal is input into the first partial model. The pre-processing includes, but is not limited to, data conversion, interpolation, sampling, filtering, noise reduction, interference removal, etc. The data conversion processing includes, but is not limited to, data dimension conversion processing, data scale conversion processing. The data conversion processing is, for example, complex to real processing.

[0150] In step S203, the terminal 101 sends the first information to the network device 102.

[0151] In some embodiments, the network device 102 receives the first information.

[0152] In some embodiments, the first information can be channel state data reported by the terminal. Alternatively, the first information can be related data reported by the terminal for estimating the channel state information.

[0153] In some embodiments, the name of the first information is not limited, which is, for example, output data or signal of the first partial model, input data or signal of the second partial model, channel state data, etc.

[0154] In some embodiments, the first information is at least one of the following:

[0155] N bits, N being an integer greater than or equal to 1;

[0156] M complex numbers, M being an integer greater than or equal to 1;

[0157] K real numbers, K being an integer greater than or equal to 1.

[0158] For example, the first information is a binary bit stream of N bits. By modulating the binary bit stream, a modulated signal can be obtained.

[0159] For example, the first information is a signal, which can be a modulated signal after modulating a binary bit stream. The signal can be expressed in complex form, i.e., the first information can be N complex numbers. The complex number includes a real part and an imaginary part, or the complex number can be understood as including the amplitude and phase of the signal. Similarly, the signal can be expressed in real form, i.e., the first information can be K real numbers.

[0160] In some embodiments, the implementation of the terminal sending the first information to the network device can be: if the first information is a binary bit stream, the first information can be modulated, and the modulated signal can be sent.

[0161] In some embodiments, one of the functions of the first partial model is modulation, and the implementation of the terminal sending the first information to the network device can be: if the first information is a modulated signal output by the first partial model, the modulated signal can be sent.

[0162] In step S204, the network device 102 inputs the first information into the second partial model to obtain the first channel state information output by the second partial model.

[0163] In some embodiments, the first information is used by the network device to determine the first channel state information. For example, the first information is used by the network device as an input of the second partial model to obtain the first channel state information output by the second partial model.

[0164] In some embodiments, the first channel state information comprises first full channel information and / or a first eigenvector corresponding to the first full channel information.

[0165] The first full channel information comprises but is not limited to a channel matrix. The channel matrix can describe a channel through which a signal passes during transmission and contains characteristics and influences of the channel, i.e., the channel matrix provides information about signal propagation, attenuation, and interference, etc.

[0166] The first eigenvector can be referred to as a first channel eigenvector. For example, the first eigenvector can be a spatial domain vector (SD). The spatial domain vector can also be referred to as a spatial domain component vector, a beam vector, a spatial domain beam basis vector, a spatial domain basis vector, etc. It should be noted that each spatial domain vector corresponds to a transmit beam of the transmitting end device. For example, the spatial domain vector can be, but is not limited to, a column vector of a two-dimensional DFT matrix or a column vector of an oversampled two-dimensional DFT matrix, i.e., the spatial domain basis vector can be a two-dimensional DFT vector. The two-dimensional DFT vector can be generally used to describe a beam superimposed by a horizontal beam and a vertical beam. For example, the spatial domain basis vector can also be an eigenvector determined according to a channel spatial domain statistical characteristic, for example, an eigenvector obtained by singular value decomposition (SVD) of the channel. For example, an eigenvector obtained by eigenvalue decomposition of a covariance matrix of the channel. Of course, the present application is not limited thereto.

[0167] In some embodiments, the second partial model is used to obtain the first channel state information according to the first information.

[0168] In some embodiments, the second partial model can be an AI model or an ML model, and the present disclosure does not make specific limitations thereon. Moreover, the name of the second partial model is not limited, which is, for example, a CSI recovery model, a second sub-model, a decoding model, etc.

[0169] In some embodiments, the first information is an input of the second partial model. By inputting the first information into the second partial model, the first channel state information output by the second partial model can be obtained, i.e., the first channel state information is the output of the second partial model.

[0170] In some embodiments, the first partial model and the second partial model are components of a bilateral model, i.e., the bilateral model includes the first partial model on the terminal side and the second partial model on the network device side. The bilateral model is different from the bilateral AI / ML model in FIG. 1C, and the bilateral AI / ML model can be implemented using a commonly used model structure such as a convolutional neural network, a transformer, etc. The specific structure of the bilateral AI / ML model is not limited in the present disclosure. The bilateral model in subsequent embodiments of the present disclosure refers to a model different from FIG. 1C.

[0171] In some embodiments, assuming that the first information is a binary bit stream, the first partial model can complete channel estimation based on the first received signal to obtain CSI data, and for the CSI data obtained by channel estimation, the CSI compression processing and the channel encoding processing can be combined, and then the encoded binary bit stream data (i.e., the first information) is output. For example, referring to FIG. 2B, the first partial model can implement the following three functions:

[0172] Channel estimation function, for obtaining CSI data by channel estimation according to the first received signal.

[0173] CSI compression and channel encoding function, for performing data compression and channel encoding to obtain the first information according to the CSI data. This function includes two sub-functions of data compression and channel encoding, i.e., while reducing the amount of CSI data, the ability to resist interference, fading, etc. is improved by encoding the data to achieve safer data transmission.

[0174] Modulation function, the data output after CSI compression and channel encoding can be directly modulated for subsequent processing and then transmitted.

[0175] In some embodiments, any function shown in FIG. 2B can be an optional function of the first partial model. For example, the modulation function is an optional function of the first partial model. Referring to FIG. 2C, the first partial model can include the channel estimation function, the CSI compression function, and the channel encoding function.

[0176] Correspondingly, if the first information output by the first part model is binary bit stream data, the second part model is configured to complete channel decoding based on the binary bit stream data, and simultaneously complete data reconstruction to obtain the first channel state information. That is, the second part model includes channel decoding and CSI reconstruction functions, and the second part model can further include demodulation functions. The channel decoding and CSI reconstruction functions are configured to output the first channel state information according to the binary bit stream data (i.e., the first information). The functions of the second part model can be understood as inverse operations (such as decompression) of CSI compression and inverse operations (such as decoding) of channel encoding in the first part model, so as to obtain CSI data close to that obtained by channel estimation in the first part model.

[0177] In some embodiments, any one of the demodulation function, the channel decoding function, and the CSI reconstruction function in the second part model can be an optional function of the second part model. For example, the second part model does not include the CSI reconstruction function.

[0178] For example, the first part model includes the channel estimation function but does not include the CSI compression and channel encoding function, and correspondingly, the second part model can not include the CSI reconstruction function.

[0179] For example, the first part model includes the channel estimation function, the CSI compression and channel encoding function, and correspondingly, the second part model can include the CSI reconstruction function.

[0180] In some possible implementation manners, the first part model does not include the channel estimation function, and the second part model includes the channel estimation function.

[0181] In some embodiments, the input of the first part model is a first received signal or a signal obtained by preprocessing the first received signal. Optionally, the first received signal refers to a received signal corresponding to all or part of frequency domain / time domain units / CDM groups, or a received signal received by the terminal after being transmitted from all or part of antenna ports of the sending end. The output of the first part model can be a binary bit stream. The input of the second part model can be a binary bit stream after channel decoding and demodulation. The output of the second part model can be first full channel information or a first channel feature vector.

[0182] In some embodiments, the bilateral AI / ML model including the first part model on the terminal side and the second part model on the network device side is trained in the following manner:

[0183] According to the sample data set, the first part model to be trained, the channel model, and the second part model to be trained are collaboratively trained to obtain a trained first part model and a trained second part model. The channel model is used to simulate a wireless channel through which the first information passes. The wireless channel through which the first information passes is a channel through which the terminal sends the first information to the network device.

[0184] The bilateral AI / ML model can be trained on a network device, a terminal, or another electronic device.

[0185] For example, during the training process, the first part model to be trained, the channel model, and the second part model to be trained are collaboratively trained in the structure shown in FIG. 2D or FIG. 2E. In some embodiments, assuming that the first information output by the first part model is binary bit stream data, the binary bit stream data is input into the channel model for processing, and the output of the channel model is input into the second part model.

[0186] The channel model is used to simulate the characteristics of a wireless channel during the training process, so that the two part models, the first part model and the second part model, are more consistent with the channel conditions in actual system applications. For example, in the presence of additive white Gaussian noise (AWGN), a binary symmetric channel network can be deployed in the channel model, which introduces errors in feedback bits according to the signal-to-noise ratio (SINR) of the channel; or a feedforward neural network can be deployed in the channel model, which inputs signal power, signal-to-noise ratio, and other features and outputs the signal after mixing noise.

[0187] In some embodiments, the sample data set includes input sample data and output sample data corresponding to the input sample data. The input sample data is a received signal, and the output sample data is channel state information.

[0188] Optionally, according to the sample data set, the first part model to be trained, the channel model, and the second part model to be trained are collaboratively trained to obtain a trained first part model and a trained second part model. The implementation of the first part model to be trained, the channel model, and the second part model to be trained is to perform at least one first process until a training model convergence condition is met to obtain the trained first part model and the trained second part model. The first process includes: inputting input sample data into the first part model; inputting the output data of the first part model into the channel model; inputting the output data of the channel model into the second part model; and adjusting the model parameters of at least one of the first part model, the channel model, and the second part model according to the error between the output data of the second part model and the output sample data.

[0189] Optionally, the first partial model to be trained, the channel model, and the second partial model to be trained are collaboratively trained according to the sample data set, and an embodiment of obtaining the trained first partial model and the trained second partial model is: inputting the input sample data into the first partial model. Inputting the output data of the first partial model into the channel model. Inputting the output data of the channel model into the second partial model. Adjusting the model parameters of at least one of the first partial model, the channel model, and the second partial model according to the error between the output data of the second partial model and the output sample data. Then, the step of inputting the input sample data into the first partial model is re-executed until the training model convergence condition is met, and the trained first partial model and the second partial model are obtained.

[0190] In some embodiments, the training model convergence condition includes at least one of the following:

[0191] The number of iterations of the training model, i.e., the number of executions of the first process, is greater than or equal to a first threshold value;

[0192] The loss function error of the training model is less than or equal to a second threshold value.

[0193] In some embodiments, the error between the output data of the second partial model and the output sample data can be a mean squared error (MSE). The mean squared error can be used as a loss function.

[0194] In some embodiments, during the model training process, methods such as stochastic gradient descent (SGD) can be used to update the model parameters to reduce the model training loss, and the training is stopped when the model loss is reduced to a certain value or the model training period reaches a certain number of times.

[0195] In some embodiments, after obtaining the trained bilateral model, the bilateral model can be tested. In the model testing process, the test received signal in the test data sample is input into the bilateral model, the output data of the bilateral model is obtained, and the error between the output data of the bilateral model and the corresponding target output data in the test data sample is calculated. The error is the main accuracy index for evaluating the effect of the bilateral model, and the smaller the error, the better the effect of the bilateral model.

[0196] In some embodiments, the sample data set includes measured sample data and / or synthetic sample data.

[0197] In some embodiments, the measured data in the real environment can be taken as measured sample data. For example, the network side transmits a pilot signal through all the transmitting antenna ports and on all the frequency domain units, and the network side can transmit the pilot signal at a regular power or at a greater transmitting power. The terminal receives the pilot signal, takes the received signal as data set 1 (representing the model input data set), and takes the full channel information or the eigenvector estimated according to the received signal as data set 2 (representing the model output data set).

[0198] In some embodiments, data set 3 and data set 4 similar to the above data set 1 and data set 2 can be generated by a simulation platform or Ray tracing based on specific simulation assumptions. Data set 4 can be regenerated by modeling phase discontinuity or antenna calibration error, etc.

[0199] In some embodiments, the above data set 1, data set 2, data set 3, and data set 4 are mixed in a certain proportion to generate a sample data set for bilateral model training.

[0200] In some embodiments, the sample data set includes at least one of the following:

[0201] The second received signal and the corresponding second channel state information, wherein the second received signal is taken as the input sample data of the sample data set, and the second channel state information is taken as the output sample data of the sample data set;

[0202] The third received signal and the corresponding third channel state information, wherein the third received signal is taken as the input sample data of the sample data set, and the third channel state information is taken as the output sample data of the sample data set.

[0203] Optionally, the second received signal is a signal received by the terminal in a first frequency domain range, and the second channel state information includes second full channel information corresponding to a second frequency domain range and / or a second eigenvector corresponding to the second full channel information, wherein the second frequency domain range at least includes the first frequency domain range. For example, assuming that the first frequency domain range is RB 1 (Resource Block 1), RB 3, RB 5, RB 7, RB 9, RB 11, RB 13, and RB 15, the second frequency domain range can be RB 1-RB 16.

[0204] Optionally, the interval between two adjacent frequency domain positions for receiving the second received signal is preset or indicated by the first device, and the first device includes the terminal and / or the network device.

[0205] Optionally, the third received signal is a signal received by the terminal after being transmitted by the network device from an antenna port in the first antenna port range and passing through a channel, and the third channel state information includes third full channel information corresponding to the second antenna port range and / or a third eigenvector corresponding to the third full channel state information, wherein the second antenna port range at least includes the first antenna port range.

[0206] Optionally, the interval between the two adjacent spatial domain positions for receiving the third received signal is pre-configured or indicated by the first device, and the first device includes the terminal and / or the network device.

[0207] In some embodiments, the number of sample data sets is a plurality, and the plurality of sample data sets correspond to a plurality of consecutive time points. By training the bilateral model through the plurality of sample data sets of the plurality of consecutive time points, the bilateral model can learn the features of the time domain correlation, so that the bilateral model can be applied to a channel estimation scenario based on the time domain correlation, and the robustness of the bilateral model is improved.

[0208] In some embodiments, the names of information and the like are not limited to the names described in the embodiments, and the terms of “information”, “message”, “signal”, “signaling”, “report”, “configuration”, “indication”, “instruction”, “command”, “channel”, “parameter”, “domain”, “field”, “symbol”, “symbol”, “codebook”, “codeword”, “codepoint”, “bit”, “data”, “program”, “chip”, and the like can be replaced with each other.

[0209] In some embodiments, the terms of “downlink”, “downlink”, “physical downlink”, and the like can be replaced with each other.

[0210] In some embodiments, the terms of “synchronization signal (SS)”, “synchronization signal block (SSB)”, “reference signal (RS)”, “pilot”, “pilot signal”, and the like can be replaced with each other.

[0211] In some embodiments, the terms “moment”, “point in time”, “time”, “time position” and the like can be replaced by each other, and the terms “duration”, “time period”, “time window”, “window”, “time” and the like can be replaced by each other.

[0212] In some embodiments, the terms “resource block (RB)”, “physical resource block (PRB)”, “sub-carrier group (SCG)”, “resource element group (REG)”, “PRB pair”, “RB pair”, “resource element (RE)”, “sub-carrier” and the like can be replaced by each other.

[0213] In some embodiments, the terms “precoding”, “precoder”, “weight”, “precoding weight”, “quasi-co-location (QCL)”, “transmission configuration indication (TCI) state”, “spatial relation”, “spatial domain filter”, “transmission power”, “phase rotation”, “antenna port”, “antenna port group”, “layer”, “the number of layers”, “rank”, “resource”, “resource set”, “resource group”, “beam”, “beam width”, “beam angular degree”, “antenna”, “antenna element”, “panel” and the like can be replaced by each other.

[0214] In some embodiments, "acquire", "obtain", "get", "receive", "transmit", "bidirectional transmission", "send and / or receive" can be replaced with each other, which can be interpreted as receiving from other subjects, acquiring from protocols, acquiring from higher layers, obtaining by self-processing, autonomously implementing, and the like.

[0215] In some embodiments, the terms "send", "transmit", "report", "issue", "transmit", "bidirectional transmission", "send and / or receive" can be replaced with each other.

[0216] In some embodiments, the terms "certain", "preset", "preset", "set", "indicated", "certain", "arbitrary", "first" and the like can be replaced with each other. "Certain A", "preset A", "preset A", "set A", "indicated A", "certain A", "arbitrary A", "first A" can be interpreted as A specified in advance in protocols and the like, A obtained by setting, configuration, or indication, and the like, A specific, A certain, A arbitrary, or A first, but not limited thereto.

[0217] The communication method related to the embodiments of the present disclosure can include at least one of steps S201-S204. For example, steps S201 and S202 can be implemented as independent embodiments, and steps S202 and S204 can be implemented as independent embodiments, but not limited thereto.

[0218] In some embodiments, the order of any two steps in steps S201-S204 can be exchanged or executed simultaneously.

[0219] In some embodiments, steps S203 and S204 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0220] In some embodiments, steps S201 and S203 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0221] In some embodiments, other optional implementations described before or after the description corresponding to FIG. 2A can be referred to.

[0222] FIG. 3A is a flow diagram of a communication method according to an embodiment of the present disclosure. As shown in FIG. 3A, the embodiments of the present disclosure relate to a communication method executed by a terminal side, and the above method comprises:

[0223] Step S3101, receiving a first received signal.

[0224] The optional implementation of step S3101 can refer to the optional implementation of step S201 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0225] In some embodiments, the terminal 101 can receive the first pilot signal sent by the network device to obtain the first received signal, but is not limited thereto, and the terminal can also receive the first pilot signal sent by other subjects to obtain the first received signal.

[0226] In step S3102, the first received signal is input into the first part model to obtain the first information.

[0227] The optional implementation of step S3102 can refer to the optional implementation of step S202 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0228] In step S3103, the first information is sent.

[0229] The optional implementation of step S3103 can refer to the optional implementation of step S203 in FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0230] In some embodiments, the terminal 101 can send the first information to the network device 102, but is not limited thereto, and can also send the first information to other subjects.

[0231] The communication method related to the embodiments of the present disclosure can include at least one of steps S3101-S3103. For example, step S3101 can be implemented as an independent embodiment, step S3102 can be implemented as an independent embodiment, and step S3103 can be implemented as an independent embodiment, but is not limited thereto.

[0232] In some embodiments, the order between any two of steps S3101-S3103 can be exchanged or executed simultaneously.

[0233] In some embodiments, steps S3101 and S3102 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0234] In some embodiments, steps S3101 and S3103 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0235] In some embodiments, steps S3102 and S3103 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0236] FIG. 3B is a flow diagram illustrating a model training according to some embodiments of the present disclosure. As shown in FIG. 3B, some embodiments of the present disclosure relate to a communication method, which is performed by a terminal side, and the above method comprises the following steps.

[0237] In step S3201, input sample data is input into the first part model.

[0238] The optional implementation of step S3201 can refer to the optional implementation of the bilateral model training in step S204 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0239] In step S3202, the output data of the first part model is input into the channel model.

[0240] The optional implementation of step S3202 can refer to the optional implementation of the bilateral model training in step S204 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0241] In step S3203, the output data of the channel model is input into the second part model.

[0242] The optional implementation of step S3203 can refer to the optional implementation of the bilateral model training in step S204 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0243] In step S3204, the parameters of the bilateral model are updated according to the error between the output data of the second part model and the output sample data.

[0244] The optional implementation of step S3204 can refer to the optional implementation of the bilateral model training in step S204 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0245] The communication method related to some embodiments of the present disclosure can comprise at least one of steps S3201-S3204. For example, step S3201 can be implemented as an independent embodiment, and step S3204 can be implemented as an independent embodiment, but is not limited thereto.

[0246] FIG. 4 is a flow diagram illustrating a communication method according to some embodiments of the present disclosure. As shown in FIG. 4, some embodiments of the present disclosure relate to a communication method, which is performed by a network device side, and the above method comprises the following steps.

[0247] In step S4101, first information is received.

[0248] The optional implementation of step S4101 can refer to the optional implementation of step S203 of FIG. 2 and other associated parts in the embodiments related to FIG. 2, which will not be repeated here.

[0249] In some embodiments, the network device 102 receives the first information sent by the terminal 101, but is not limited thereto, and can also receive the first information sent by other subjects.

[0250] Step S4102: inputting the first information into the second part model to obtain first channel state information.

[0251] The optional implementation of step S4102 can refer to the optional implementation of step S204 in FIG.2 and other associated parts in the embodiments involved in FIG.2, which will not be repeated here.

[0252] The communication method involved in the embodiments of the present disclosure can include at least one of step S4101 and step S4102. For example, step S4101 can be implemented as an independent embodiment, and step S4102 can be implemented as an independent embodiment, but is not limited thereto.

[0253] In some embodiments, step S4101 and step S4102 can be exchanged in order or executed simultaneously.

[0254] In some embodiments, step S4101 is optional, and can be omitted or replaced in different embodiments.

[0255] In some embodiments, step S4102 is optional, and can be omitted or replaced in different embodiments.

[0256] In some embodiments, the method of training the bilateral model from the network device side includes the following steps:

[0257] Step S4201: inputting sample data into the first part model.

[0258] The optional implementation of step S4201 can refer to the optional implementation of step S204 in FIG.2 and other associated parts in the embodiments involved in FIG.2, which will not be repeated here.

[0259] Step S4202: inputting the output data of the first part model into the channel model.

[0260] The optional implementation of step S4202 can refer to the optional implementation of step S204 in FIG.2 and other associated parts in the embodiments involved in FIG.2, which will not be repeated here.

[0261] Step S4203: inputting the output data of the channel model into the second part model.

[0262] The optional implementation of step S4203 can refer to the optional implementation of the bilateral model training under step S204 of FIG. 2 and other associated parts in the embodiments involved in FIG. 2, which will not be repeated here.

[0263] In step S4204, the parameters of the bilateral model are updated according to the error between the output data of the second part model and the output sample data.

[0264] The optional implementation of step S4204 can refer to the optional implementation of the bilateral model training under step S204 of FIG. 2 and other associated parts in the embodiments involved in FIG. 2, which will not be repeated here.

[0265] The communication method involved in the embodiments of the present disclosure can include at least one of steps S4201-S4204. For example, step S4201 can be implemented as an independent embodiment, and step S4204 can be implemented as an independent embodiment, but is not limited thereto.

[0266] In some embodiments, it is assumed that the pilot signal sent by the sending end passes through the channel to reach the receiving end, and the signal of the pilot signal passing through the channel to reach the receiving end is defined as the received signal.

[0267] In some embodiments, the AI / ML model includes a CSI compression model and a CSI recovery model. The CSI compression model is equivalent to the first part model in the foregoing embodiments, and the CSI recovery model is equivalent to the second part model in the foregoing embodiments.

[0268] Optionally, the input and output of the AI / ML model are designed as:

[0269] The input of the CSI compression model: the received signal or the signal processed from the received signal. Optionally, the received signal refers to the received signal corresponding to part of the frequency domain / time domain unit / CDM group, or the received signal received on part of the antenna port.

[0270] The output of the CSI compression model: an N-bit binary bit stream, N is greater than or equal to 1.

[0271] The input of the CSI recovery model: the binary bit stream after channel decoding and demodulation.

[0272] The output of the CSI recovery model: full channel information or channel feature vector.

[0273] In some embodiments, the data set used to train the AI / ML model includes the following data set features:

[0274] Data feature 1 (CSI compression feedback based on frequency domain correlation):

[0275] Dataset 1: received signals within a bandwidth range (there are multiple signals within the bandwidth) at a certain time. The frequency domain unit interval between adjacent received signals is f, which can be a subcarrier or a PRB or a subband, etc. The value of f can be determined by predefinition, network configuration or terminal reporting.

[0276] Dataset 2: full channel information corresponding to the frequency domain unit where the received signal is located or the eigenvector corresponding to the full channel information. Optionally, the corresponding full channel information or the eigenvector corresponding to the full channel information on part or all of the frequency domain units.

[0277] Data feature 2 (CSI compression feedback based on spatial correlation):

[0278] Dataset 1: received signals corresponding to part of the transmit antenna ports or all of the transmit antenna ports at a certain time. The part of the transmit antenna ports can be determined by network configuration or predefined transmit antenna port pattern. If there are multiple transmit antenna port patterns, the network side needs to indicate the transmit antenna port pattern adopted by the network through signaling.

[0279] Dataset 2: full channel information corresponding to part or all of the transmit antenna ports or the eigenvector corresponding to the full channel information.

[0280] Data feature 3: combination of data features 1 and 2.

[0281] Based on the above dataset 1 and dataset 2, a bilateral AI / ML model is trained.

[0282] In some embodiments, the dataset for training the AI / ML model is obtained by at least one of the following ways:

[0283] Way 1: measured data from the field environment. The network side transmits pilot signals through all transmit antenna ports, and there are pilot signals transmitted on all frequency domain units. The network side can transmit the pilot signals at a regular power or at a larger transmission power. The pilot signals received by the terminal are used as dataset 1, and the eigenvector of the full channel information estimated or determined according to the received signal is used as dataset 2.

[0284] Way 2: generate dataset 1 and dataset 2 based on specific simulation assumptions through simulation platform or ray tracing. Dataset 2 can be regenerated by modeling phase discontinuity or antenna calibration error, etc.

[0285] Way 3: combine the dataset 1 and dataset 2 generated by way 1 and way 2 in a certain proportion.

[0286] In some embodiments, the design and training of the AI / ML model are as follows:

[0287] The functions that can be realized by the CSI compression model in sequence are: completing channel estimation based on the pilot received signal (i.e., the received signal) to obtain CSI, completing joint compression of CSI data and channel coding for the CSI data obtained by channel estimation, and then outputting the coded binary bit stream data, as shown in FIG. 2B. Therefore, one CSI compression model realizes the functions of the following two sub-models:

[0288] 1. Channel estimation function: input the received signal, and obtain full channel information or full channel eigenvector information through the channel estimation model.

[0289] 2. CSI compression and channel coding function: input the full channel information or full channel eigenvector information, and output the bit data after data compression and channel coding. The model can realize both data compression and channel coding functions, that is, while reducing the amount of CSI data, the ability to resist interference, fading, etc. is improved through coding processing of the data to realize safer data transmission. The data output by the CSI compression and channel coding model can be directly modulated and processed for subsequent processing and then transmitted.

[0290] The binary bit stream data output by the CSI compression model will be input into the channel model for certain processing, and the output of the model will be input into the CSI recovery model. The channel model is used to simulate the characteristics of the wireless channel in the training process, so that the CSI compression and recovery models are more consistent with the channel conditions in actual system applications.

[0291] For example, in the presence of additive white Gaussian noise (AWGN), a binary symmetric channel network can be deployed, which introduces errors in feedback bits according to the signal-to-noise ratio (SINR) of the channel; or a feedforward neural network can be deployed, which inputs signal power, signal-to-noise ratio, etc. Features, output the received signal after mixing noise.

[0292] Optionally, the functions that can be realized by the CSI recovery model are: completing channel decoding based on the received binary bit stream data, and simultaneously completing data reconstruction to obtain CSI data. Therefore, the CSI recovery model realizes the functions of channel decoding and CSI reconstruction.

[0293] For example, the channel decoding and CSI reconstruction function is: inputting the binary bit stream data output by the channel model, and outputting the full channel information or full channel eigenvector information. The model realizes the inverse operation of the CSI compression and channel coding model in the CSI compression model, and obtains CSI data close to the output of the channel estimation model.

[0294] In some embodiments, assuming a system bandwidth of 16 RBs, the gNB transmits downlink pilot signals CSI-RS to the UE, with a CSI-RS density of 0.5, i.e., f = 2 PRBs. Referring to Figure 5, at each CSI-RS transmission time and over the entire bandwidth, the UE estimates the downlink channel H on each RB based on the received signal y. Assume the gNB transmits via antenna N. t =32, UE receiving antenna N r =4. Among them, the number of CSI-RS ports is 32, and N within each RB transmitting CSI-RS... p =8 subcarriers, and N s = Transmission over 4 OFDM symbols, number of subcarriers within the bandwidth Nc = 12 * N w =192. N w Let N represent the total number of RBs. The total number of subcarriers in CSI-RS transmission is N = 64. Therefore, the data dimension of the received signal Y is: number of receiving antennas × number of pilot subcarriers × number of OFDM symbols, i.e. The channel matrix H data dimension is: number of receive antennas × number of transmit antennas × number of subcarriers within the bandwidth, i.e. The input and output data used for AI / ML training are typically real numbers. Therefore, the complex data of Y and H need to be converted to real numbers first. The corresponding data dimensions of Y and H are then Y∈R. 4×64×4×2 and H∈R 4×32×192×2 If the dataset is generated based on a simulation platform, in order to reflect the impact of noise in the real system, different SNR (signal-to-noise ratio) values ​​are set to obtain datasets of Y and H corresponding to different SNRs.

[0295] In some embodiments, assuming the gNB transmits CSI-RS to the UE at a lower density period, such as a CSI-RS density of 0.25 (f = 4pRB), as shown in Figure 6, the UE will receive pilot signals in fewer RBs. The UE can infer the channel information for the remaining RBs based on the AI / ML model, or estimate the channel information for RBs without pilot transmission using traditional interpolation algorithms, such as RB2, RB3, and RB4. Similarly, the UE can also estimate the channel information for RBs without pilot transmission using the AI / ML model or traditional algorithms, which will not be elaborated here. The channel information estimated based on the frequency and spatial domains can also be applied to AI / ML channel estimation based on time-domain correlation to improve channel estimation performance.

[0296] In some embodiments, it is assumed that the gNB transmitting antenna N t =32, UE receiving antenna N r= 4. However, the gNB only transmits 16 out of 32 ports. The 16 antenna ports can be determined by a predefinition. For example, the predefinition is to transmit only the antenna ports with odd or even indices. Or the predefinition is to transmit only the antenna ports in a single polarization direction.

[0297] Optionally, the UE side can infer the channel information of the remaining transmitting antenna ports by a traditional interpolation algorithm, or deploy a module for predicting the channel of other antenna ports on the UE side as part of the AI / ML model, and obtain the channel information of the other antenna ports by the module, so as to obtain the channel information of all transmitting antenna ports. For example, the interpolation inference function is deployed in the first part of the model.

[0298] In some embodiments, the design, training and testing of the bilateral AI / ML model include:

[0299] (1) Model design, based on the above data set acquisition method, a plurality of pilot received signals and corresponding CSI data are obtained to form a data sample, which is used as input data and output data of the bilateral AI / ML model for joint channel estimation, channel feedback and channel coding. The specific structure of the bilateral AI / ML model is not limited in the present application, and can be realized by using common model structures such as convolutional neural network and Transformer.

[0300] (2) Model training, in the model training process, the pilot received signal in the sample data is taken as the input data, and the actual output data is obtained after being processed by the CSI compression model, the channel model and the CSI recovery model in turn, the channel H in the sample data is taken as the target output data of the AI / ML model, the mean squared error (MSE) between the actual output data and the target output data of the AI / ML model is taken as the loss function, and the model parameters are updated by using methods such as stochastic gradient descent (SGD) to reduce the model training loss in the model training process, and the training is stopped when the model loss is reduced to a certain value or the model training period reaches a certain number of times.

[0301] (3) Model testing, based on the AI / ML model obtained in (2), the pilot received signal in the test data sample is input in the model testing process, the output data of the AI / ML model is obtained, and the error between the model output data and the corresponding target output data in the test set is calculated as the main accuracy index for evaluating the effect of the AI / ML model, and the smaller the error is, the better the channel estimation effect of the AI / ML model is.

[0302] The bilateral model design for joint channel estimation, CSI compression and channel coding proposed in the embodiments of the present disclosure can reduce the complexity of model design and model management.

[0303] In the embodiments of the present disclosure, part or all of the steps, and optional implementation manners thereof, can be combined with part or all of the steps in other embodiments, or combined with optional implementation manners of other embodiments.

[0304] The embodiments of the present disclosure also propose a device for implementing any of the above methods, for example, a device comprising units or modules for implementing the steps performed by a terminal in any of the above methods. For another example, another device is proposed, comprising units or modules for implementing the steps performed by a network device (such as an access network device, a core network function node, a core network device, etc.) in any of the above methods.

[0305] It should be understood that the division of each unit or module in the above device is only a logical function division, and all or part of them can be integrated into one physical entity, or can be physically separated. In addition, the units or modules in the device can be implemented in the form of processor calling software: for example, the device includes a processor, the processor is connected with a memory, the memory stores instructions, and the processor calls the instructions stored in the memory to implement any of the above methods or the functions of each unit or module of the device, wherein the processor is, for example, a general processor such as a central processing unit (CPU) or a microprocessor, and the memory is a memory within the device or a memory outside the device. Alternatively, the units or modules in the device can be implemented in the form of hardware circuit, and the functions of part or all of the units or modules can be implemented by designing the hardware circuit, and the hardware circuit can be understood as one or more processors; for example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC), and the functions of part or all of the units or modules are implemented by designing the logical relationship of elements in the circuit; for another example, in another implementation, the hardware circuit is a programmable logic device (PLD), and a field programmable gate array (FPGA) is taken as an example, which can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured by a configuration file, so as to implement the functions of part or all of the units or modules. All units or modules of the above device can be implemented in the form of processor calling software, or all units or modules can be implemented in the form of hardware circuit, or part of the units or modules can be implemented in the form of processor calling software, and the remaining part can be implemented in the form of hardware circuit.

[0306] In the embodiments of the present disclosure, the processor is a circuit with signal processing capability. In one implementation, the processor can be a circuit with instruction reading and running capability, such as a central processing unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a microprocessor), a digital signal processor (DSP), or the like. In another implementation, the processor can implement certain functions through a logical relationship of a hardware circuit, and the logical relationship of the hardware circuit is fixed or can be reconfigured. For example, the processor is a hardware circuit implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as an FPGA. In the reconfigurable hardware circuit, the processor loads a configuration document to implement the hardware circuit configuration. It can be understood that the processor loads instructions to implement the functions of the above part or all units or modules. In addition, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), and the like.

[0307] FIG. 7A is a schematic diagram of a structure of a terminal according to an embodiment of the present disclosure. As shown in FIG. 7A, the terminal 7100 can include at least one of a transceiver module 7101, a processing module 7102, and the like. In some embodiments, the transceiver module 7101 is configured to receive a first received signal, where the first received signal is a signal received by the terminal after a first pilot signal transmitted by a network device passes through a channel; the processing module 7102 is configured to input the first received signal into a first part model trained to obtain first information output by the first part model; the transceiver module 7101 is further configured to send the first information to the network device, where the first information is used by the network device as an input of a second part model trained to obtain first channel state information output by the second part model. Optionally, the transceiver module 7101 is configured to perform at least one of the communication steps (for example, steps S201 and S203, but not limited to) performed by the terminal 101 in any of the methods described above, and details are not described herein again. Optionally, the processing module 7102 is configured to perform at least one of the other steps (for example, steps S202 and S204, but not limited to) performed by the terminal 101 in any of the methods described above, and details are not described herein again.

[0308] FIG. 7B is a schematic diagram of a structure of a network device according to an embodiment of the present disclosure. As shown in FIG. 7B, the network device 7200 can include at least one of a transceiver module 7201, a processing module 7202, and the like. In some embodiments, the transceiver module 7201 is configured to receive first information sent by a terminal, where the first information is obtained by the terminal after a first received signal input into a first part model trained, and the first received signal is a signal received by the terminal after a first pilot signal transmitted by the network device passes through a channel; the processing module 7202 is configured to input the first information into a second part model trained to obtain first channel state information output by the second part model. Optionally, the transceiver module 7201 is configured to perform at least one of the communication steps (for example, steps S201 and S203, but not limited to) performed by the network device 102 in any of the methods described above, and details are not described herein again. Optionally, the processing module 7202 is configured to perform at least one of the other steps (for example, steps S202 and S204, but not limited to) performed by the network device 102 in any of the methods described above, and details are not described herein again.

[0309] In some embodiments, the transceiver module can include a sending module and / or a receiving module, and the sending module and the receiving module can be separate or integrated together. Optionally, the transceiver module can be replaced by a transceiver.

[0310] In some embodiments, the processing module can be one module, or can include multiple sub-modules. Optionally, the multiple sub-modules perform all or part of the steps required to be performed by the processing module, respectively. Optionally, the processing module can be mutually replaced with the processor.

[0311] FIG. 8A is a structural schematic diagram of a communication device 8100 according to an embodiment of the present disclosure. The communication device 8100 can be a network device (for example, an access network device, a core network device, etc.), a terminal (for example, a user equipment, etc.), a chip, a chip system, or a processor supporting the network device to implement any of the above methods, or a chip, a chip system, or a processor supporting the terminal to implement any of the above methods. The communication device 8100 can be used to implement the methods described in the above method embodiments, and details can be referred to the descriptions in the above method embodiments.

[0312] As shown in FIG. 8A, the communication device 8100 includes one or more processors 8101. The processor 8101 can be a general purpose processor or a special purpose processor, for example, a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data, and the central processing unit can be used to control the communication device (for example, a base station, a baseband chip, a terminal device, a terminal device chip, a DU or a CU, etc.), execute programs, and process data of the programs. Optionally, the communication device 8100 is used to execute any of the above methods. Optionally, the one or more processors 8101 are used to call instructions to enable the communication device 8100 to execute any of the above methods.

[0313] In some embodiments, the communication device 8100 further includes one or more transceivers 8102. When the communication device 8100 includes the one or more transceivers 8102, the transceiver 8102 performs at least one of the communication steps (for example, step S201, step S203, but not limited to this) in the above method, and the processor 8101 performs at least one of the other steps (for example, step S202, step S204, but not limited to this). In an optional embodiment, the transceiver can include a receiver and / or a transmitter, which can be separate or integrated together. Optionally, the terms of transceiver, transceiving unit, transceiver, transceiving circuit, interface circuit, interface, etc. can be mutually replaced, and the terms of transmitter, transmitting unit, transmitter, transmitting circuit, etc. can be mutually replaced, and the terms of receiver, receiving unit, receiver, receiving circuit, etc. can be mutually replaced.

[0314] In some embodiments, the communication device 8100 further includes one or more memories 8103 for storing data. Alternatively, all or part of the memories 8103 can be external to the communication device 8100. In optional embodiments, the communication device 8100 can include one or more interface circuits 8104. Optionally, the interface circuit 8104 is connected to the memory 8103, and the interface circuit 8104 can be used to receive data from the memory 8103 or other devices, and can be used to send data to the memory 8103 or other devices. For example, the interface circuit 8104 can read data stored in the memory 8103 and send the data to the processor 8101.

[0315] The communication device 8100 described in the above embodiments can be a network device or a terminal, but the scope of the communication device 8100 described in the present disclosure is not limited thereto, and the structure of the communication device 8100 can not be limited by Figure 8A. The communication device can be a standalone device or can be part of a larger device. For example, the communication device can be: 1) a standalone integrated circuit (IC), or a chip, or a chip system or subsystem; (2) a set of one or more ICs, which can optionally include a storage component for storing data, programs; (3) an ASIC, such as a modem; (4) a module that can be embedded in other devices; (5) a receiver, a terminal device, a smart terminal device, a cellular phone, a wireless device, a handset, a mobile unit, a vehicle-mounted device, a network device, a cloud device, an artificial intelligence device, etc.; (6) other devices, etc.

[0316] Figure 8B is a structural schematic diagram of a chip 8200 according to an embodiment of the present disclosure. For the case where the communication device 8100 is a chip or a chip system, the structural schematic diagram of the chip 8200 shown in Figure 8B can be referred to, but is not limited thereto.

[0317] The chip 8200 includes one or more processors 8201. The chip 8200 is configured to execute any of the above methods.

[0318] In some embodiments, the chip 8200 further includes one or more interface circuits 8202. Optionally, the terms interface circuit, interface, transceiver pin, etc. can be replaced by each other. In some embodiments, the chip 8200 further includes one or more memories 8203 for storing data. Optionally, all or part of the memories 8203 can be external to the chip 8200. Optionally, the interface circuit 8202 is connected to the memory 8203, and the interface circuit 8202 can be used to receive data from the memory 8203 or other devices, and the interface circuit 8202 can be used to send data to the memory 8203 or other devices. For example, the interface circuit 8202 can read data stored in the memory 8203 and send the data to the processor 8201.

[0319] In some embodiments, the interface circuit 8202 performs at least one of the communication steps (for example, step S201, step S203, but not limited thereto) of transmitting and / or receiving and the like in the above method. The interface circuit 8202 performing the communication steps of transmitting and / or receiving and the like in the above method refers to, for example, the interface circuit 8202 performing data interaction between the processor 8201, the chip 8200, the memory 8203, or the transceiver device. In some embodiments, the processor 8201 performs at least one of the other steps (for example, step S202, step S204, but not limited thereto).

[0320] The various modules and / or devices described in each of the embodiments of the virtual device, the physical device, the chip, and the like can be combined or separated as the case can be. Alternatively, part or all of the steps can also be performed by a plurality of modules and / or devices in cooperation, which is not limited here.

[0321] The disclosure also proposes a storage medium, and the above storage medium stores instructions, which, when executed on the communication device 8100, cause the communication device 8100 to perform any of the above methods. Alternatively, the above storage medium is an electronic storage medium. Alternatively, the above storage medium is a computer readable storage medium, but is not limited thereto, and it can also be a storage medium readable by other devices. Alternatively, the above storage medium can be a non-transitory storage medium, but is not limited thereto, and it can also be a transitory storage medium.

[0322] The disclosure also proposes a program product, and the above program product is executed by the communication device 8100, so that the communication device 8100 performs any of the above methods. Alternatively, the above program product is a computer program product.

[0323] The disclosure also proposes a computer program, which, when executed on a computer, causes the computer to perform any of the above methods.

Claims

A communication method characterized by comprising: The method is performed by a terminal and comprises: receiving a first received signal, the first received signal being a signal received by the terminal after a first pilot signal transmitted by a network device is transmitted through a channel; inputting the first received signal into a first part model trained, to obtain first information output by the first part model; sending the first information to the network device, the first information being used by the network device as an input of a second part model trained, to obtain first channel state information output by the second part model. The method of claim 1, wherein The first part model and the second part model are components of a double-sided model. The method according to claim 2, characterized in that The double-sided model is trained in the following manner: According to a sample data set, the first part model to be trained, a channel model, and the second part model to be trained are collaboratively trained to obtain the first part model trained and the second part model trained. The channel model is used to simulate a wireless channel through which the first information passes. The method according to claim 3, characterized in that The sample data set includes input sample data and output sample data corresponding to the input sample data. According to the sample data set, the first part model to be trained, the channel model, and the second part model to be trained are collaboratively trained to obtain the first part model trained and the second part model trained, comprising: performing at least one first process until a training model convergence condition is met, to obtain the first part model trained and the second part model trained. The first process comprises: inputting the input sample data into the first part model; inputting output data of the first part model into the channel model; inputting output data of the channel model into the second part model; adjusting model parameters of at least one of the first part model, the channel model, and the second part model according to an error between output data of the second part model and the output sample data. The method according to claim 4, characterized in that The training model convergence condition comprises at least one of the following: a number of times of performing the first process is greater than or equal to a first threshold value; a loss function error of a training model is less than or equal to a second threshold value. The method according to any one of claims 3-5, characterized in that The sample data set comprises at least one of the following: a second received signal and corresponding second channel state information, wherein the second received signal is a signal received by the terminal in a first frequency domain range; a third received signal and corresponding third channel state information, wherein the third received signal is a signal received by the terminal after the network device transmits a signal from an antenna port in a first antenna port range through a channel. The method according to claim 6, characterized in that The second channel state information comprises second full channel information corresponding to a second frequency domain range and / or a second feature vector corresponding to the second full channel information, wherein the second frequency domain range at least includes the first frequency domain range. The method according to claim 6 or 7, characterized in that The third channel state information includes third full channel information corresponding to a second antenna port range and / or a third eigenvector corresponding to the third full channel state information, wherein the second antenna port range at least includes the first antenna port range. The method according to any one of claims 6-8, characterized in that The interval between adjacent two frequency domain positions for receiving the second receiving signal is preset or indicated by a first device, and the first device includes the terminal and / or the network device. The method according to any one of claims 6-9, characterized in that The interval between adjacent two spatial domain positions for receiving the third receiving signal is preset or indicated by a first device, and the first device includes the terminal and / or the network device. A communication method characterized by comprising: The method is performed by a network device, and the method comprises: receiving first information sent by a terminal, wherein the first information is obtained by inputting a first receiving signal received by the terminal into a trained first part model, and the first receiving signal is a signal received by the terminal after a first pilot signal sent by the network device is transmitted through a channel; inputting the first information into a trained second part model to obtain first channel state information output by the second part model. The method of claim 11, wherein The first part model and the second part model are components of a double-sided model. The method of claim 12, wherein The double-sided model is trained in the following manner: According to a sample data set, the first part model to be trained, a channel model, and the second part model to be trained are collaboratively trained to obtain the trained first part model and the trained second part model. The channel model is used to simulate a wireless channel through which the first information passes. The method of claim 13, wherein The sample data set includes input sample data and output sample data corresponding to the input sample data. According to a sample data set, the first part model to be trained, a channel model, and the second part model to be trained are collaboratively trained to obtain the trained first part model and the trained second part model, comprising: performing at least one first process until a trained model convergence condition is met to obtain the trained first part model and the trained second part model; The first process comprises: inputting the input sample data into the first part model; inputting output data of the first part model into the channel model; inputting output data of the channel model into the second part model; adjusting model parameters of at least one of the first part model, the channel model, and the second part model according to an error between output data of the second part model and the output sample data. The trained model convergence condition comprises at least one of the following: The method of claim 14, wherein The number of times of performing the first process is greater than or equal to a first threshold value; A loss function error of a trained model is less than or equal to a second threshold value. The sample data set comprises at least one of the following: The method according to any one of claims 13-15, characterized in that a second receiving signal and corresponding second channel state information, wherein the second receiving signal is a signal received by the terminal in a first frequency domain range; ​ a third receive signal and corresponding third channel state information, wherein the third receive signal is a signal received by the terminal after a signal transmitted by the network device from an antenna port in a first antenna port range and passing through a channel. The method of claim 16, wherein The second channel state information comprises second full channel information corresponding to a second frequency domain range and / or a second eigenvector corresponding to the second full channel information, wherein the second frequency domain range at least comprises the first frequency domain range. The method according to claim 16 or 17, characterized in that The third channel state information comprises third full channel information corresponding to a second antenna port range and / or a third eigenvector corresponding to the third full channel state information, wherein the second antenna port range at least comprises the first antenna port range. The method according to any one of claims 16-18, characterized in that An interval between two adjacent frequency domain positions for receiving the second receive signal is preset or indicated by a first device, wherein the first device comprises the terminal and / or the network device. The method according to any one of claims 16-19, characterized in that An interval between two adjacent spatial domain positions for receiving the third receive signal is preset or indicated by a first device, wherein the first device comprises the terminal and / or the network device. A terminal, characterized by comprising: comprising: a transceiver module configured to receive a first receive signal, wherein the first receive signal is a signal received by the terminal after a first pilot signal transmitted by the network device and passing through a channel; a processing module configured to input the first receive signal into a first part model, to obtain first information output by the first part model; the transceiver module is further configured to send the first information to the network device, wherein the first information is used as an input of a second part model by the network device, to obtain first channel state information output by the second part model. A network device, characterized in that comprising: a transceiver module configured to receive first information sent by the terminal, wherein the first information is obtained by inputting a first receive signal into a first part model by the terminal, and the first receive signal is a signal received by the terminal after a first pilot signal transmitted by the network device and passing through a channel; a processing module configured to input the first information into a second part model, to obtain first channel state information output by the second part model. A communication device characterized by comprising: comprising: one or more processors; a memory coupled to the processors, the memory having stored thereon executable instructions that, as a result of execution by the one or more processors, cause the communication method of any one of claims 1-20 to be performed. A communication system characterized by comprising a terminal and a network device, wherein the terminal is configured to receive a first receive signal, wherein the first receive signal is a signal received by the terminal after a first pilot signal transmitted by the network device and passing through a channel; input the first receive signal into a first part model, to obtain first information output by the first part model; and send the first information to the network device; the network device is configured to receive the first information sent by the terminal, and input the first information into a second part model, to obtain first channel state information output by the second part model. A storage medium storing instructions, the storage medium storing instructions, characterized in that, When the instructions are executed on the communication device, the communication device is caused to perform the communication method of any one of claims 1-20. A computer program product comprising computer programs and / or instructions, characterized in that, The computer program and / or instructions, when executed on the communication device, implement the communication method of any one of claims 1-20.

Images