Transmission switching method and device, communication equipment and storage medium

Abstract

The embodiment of the invention discloses a transmission switching method and device, communication equipment and a storage medium. The method comprises the following steps: a first network device sends a first message to a second network device, wherein the first message is used for requesting switching; the first message comprises model related information and model data segment related information; the first network equipment receives a second message sent by the second network equipment, wherein the second message represents switching request confirmation; the second message includes information related to processing of the model data segments.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and in particular, to a transmission switching method, apparatus, communication device, and storage medium. Background Art

[0002] Current mainstream use cases of physical layer artificial intelligence (AI) include channel state information (CSI) compression feedback and beam management. In the CSI compression feedback use case, the model is deployed on both the terminal and the base station, and in the beam management use case, the model is deployed on the terminal. Considering that model training is generally performed on the network side, how to transmit the AI model to the terminal must also be considered. Transmission based on the control plane (CP) can largely avoid significant changes to the protocol stack architecture. However, considering that the data size of the AI model is significantly larger than the existing control signaling size, how to segment the transmission of the AI model is a problem that must be faced, especially when base station handover occurs. Considering that the data volume of the AI model may be large, if all model data is retransmitted every time a handover occurs, the transmission efficiency will be severely affected. Summary of the Invention

[0003] To solve the existing technical problems, embodiments of the present invention provide a transmission switching method, apparatus, communication device, and storage medium.

[0004] To achieve the above object, the technical solution of the embodiments of the present invention is implemented as follows:

[0005] In a first aspect, an embodiment of the present invention provides a transmission switching method, which is applied to a first network device, and the method includes:

[0006] The first network device sends a first message to a second network device, and the first message is used to request a handover; the first message includes model-related information and model data segmentation-related information;

[0007] The first network device receives a second message sent by the second network device, and the second message indicates confirmation of the handover request; the second message includes information related to the processing of model data segmentation.

[0008] In the above solution, the method further includes: the first network device sends first information to the second network device, and the first information indicates the download status of the model data.

[0009] In the above solution, the model-related information includes second information, and the second information represents the features of the first model transmitted to the terminal; the model data segment-related information includes a first value and a first identifier, the first value represents the number of model data segments that have been transmitted, and the first identifier represents the existence or non-existence of a model data segment being transmitted.

[0010] In the above solution, the second information is information associated with the model type and / or the training method of the model, or the second information is a model version identifier associated with the model parameters.

[0011] In the above solution, the second message at least includes third information, and the third information represents discarding or not discarding the transmitted model data segments.

[0012] In a second aspect, an embodiment of the present invention further provides a transmission switching method, which is applied to a second network device, and the method includes:

[0013] The second network device receives a first message sent by a first network device, and the first message is used to request a handover; the first message includes model-related information and model data segment-related information;

[0014] The second network device sends a second message to the first network device, and the second message represents a handover request confirmation; the second message includes information related to the processing of model data segments.

[0015] In the above solution, the method further includes: the second network device receives first information sent by the first network device, and the first information represents the download status of the model data.

[0016] In the above solution, the model-related information includes second information, and the second information represents the features of the first model transmitted to the terminal; the model data segment-related information includes a first value and a first identifier, the first value represents the number of model data segments that have been transmitted, and the first identifier represents the existence or non-existence of a model data segment being transmitted.

[0017] In the above solution, the method further includes: the second network device retransmits the model data segments of the first model to the terminal or continues to transmit the model data segments of the first model.

[0018] In the above solution, the continuing to transmit the model data segments of the first model includes:

[0019] The second network device determines the identifier of the model data segment to continue transmission according to the first value and the first identifier, or determines the identifier of the model data segment to continue transmission according to the first identifier and the first information, and continues to transmit the model data segment of the first model to the terminal starting from the identifier of the model data segment to continue transmission.

[0020] In the above solution, multiple model data segments corresponding to the first model are pre-stored in the second network device.

[0021] In a third aspect, an embodiment of the present invention further provides a transmission switching method, which is applied to a first network device, and the method includes:

[0022] The first network device initiates data transmission of a first model and sends a model data segment of the first model to a terminal;

[0023] When the first network device makes a handover decision and determines that the terminal accesses a second network device, it continues to transmit the model data segment of the first model that has not been transmitted completely to the terminal, or sends fourth information to the terminal, where the fourth information is used to notify discarding the model data segment of the first model that has been transmitted.

[0024] In the above solution, the sending the model data segment of the first model to the terminal includes:

[0025] The first network device sends the model data segment of the first model to the terminal through a Radio Resource Control (RRC) signaling; and / or,

[0026] The continuing to transmit the model data segment of the first model that has not been transmitted completely to the terminal includes:

[0027] The first network device continues to transmit the model data segment of the first model that has not been transmitted completely to the terminal through the RRC signaling.

[0028] In the above solution, when the first network device makes a handover decision and determines that the terminal accesses a second network device, the method further includes:

[0029] The first network device sends a first message to the second network device, where the first message is used to request a handover; the first message includes model-related information and model data segment-related information;

[0030] The first network device receives a second message sent by the second network device, where the second message indicates confirmation of the handover request; the second message includes information related to the processing of the model data segment.

[0031] In the above solution, the model-related information includes second information, and the second information represents the features of the first model transmitted to the terminal; the model data segment-related information includes a first value and a first identifier, the first value represents the number of model data segments that have been transmitted, and the first identifier represents the existence or non-existence of a model data segment being transmitted; and / or,

[0032] The second message includes at least third information, and the third information represents discarding or not discarding the transmitted model data segments.

[0033] In the above solution, the second information is information associated with the model type and / or the training method of the model, or the second information is a model version identifier associated with the model parameters.

[0034] In the above solution, continuing to transmit the model data segments of the first model that have not been completely transmitted to the terminal includes: when the first identifier represents the existence of a model segment data being transmitted, the first network device continues to transmit the model data segments of the first model that have not been completely transmitted to the terminal.

[0035] In the above solution, sending fourth information to the terminal includes: when the third information represents discarding the transmitted model data segments, sending fourth information to the terminal.

[0036] In the above solution, a plurality of model data segments corresponding to the first model are pre-stored in the first network device.

[0037] In a fourth aspect, an embodiment of the present invention further provides a transmission switching method, and the method is applied to a terminal, and the method includes:

[0038] The terminal receives model data segments of a first model transmitted by a first network device;

[0039] When the terminal switches from the first network device to access a second network device, the terminal receives the model data segments of the first model that have not been completely transmitted continuously transmitted by the first network device, or receives fourth information sent by the first network device, and the fourth information is used to notify discarding the transmitted model data segments of the first model.

[0040] In the above solution, when the terminal receives the fourth information sent by the first network device, the method further includes: the terminal discards the received model data segments of the first model.

[0041] In the above solution, the method further includes: the terminal receives the model data segments of the first model retransmitted by the second network device.

[0042] In the above solution, the method further includes: the terminal receives the segmented model data of the first model continuously transmitted by the second network device.

[0043] In the above solution, the first network device and the second network device prestored therein a plurality of segmented model data corresponding to the first model.

[0044] In a fifth aspect, an embodiment of the present invention further provides a transmission switching device, which is applied to a first network device. The device includes a first communication unit, configured to send a first message to a second network device, where the first message is used to request a handover; the first message includes model-related information and segmented model data-related information; and is further configured to receive a second message sent by the second network device, where the second message indicates a handover request confirmation; the second message includes information related to the processing of the segmented model data.

[0045] In a sixth aspect, an embodiment of the present invention further provides a transmission switching device, which is applied to a second network device. The device includes a second communication unit, configured to receive a first message sent by a first network device, where the first message is used to request a handover; the first message includes model-related information and segmented model data-related information; and is further configured to send a second message to the first network device, where the second message indicates a handover request confirmation; the second message includes information related to the processing of the segmented model data.

[0046] In a seventh aspect, an embodiment of the present invention further provides a transmission switching device, which is applied to a first network device. The device includes a third communication unit, configured to initiate data transmission of a first model and send the segmented model data of the first model to a terminal; and is further configured to, when it is determined through a handover decision that the terminal accesses the second network device, continue to send the segmented model data of the first model that has not been completely transmitted to the terminal, or send fourth information to the terminal, where the fourth information is used to notify discarding the segmented model data of the first model that has been transmitted.

[0047] In an eighth aspect, an embodiment of the present invention further provides a transmission switching device, which is applied to a terminal. The device includes a fourth communication unit, configured to receive the segmented model data of a first model transmitted by a first network device; and is further configured to, when switching from the first network device to access the second network device, receive the segmented model data of the first model that has not been completely transmitted continuously sent by the first network device, or receive fourth information sent by the first network device, where the fourth information is used to notify discarding the segmented model data of the first model that has been transmitted.

[0048] In a ninth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the steps of the transmission switching method described in any one of the first to fourth aspects of the embodiments of the present invention are implemented.

[0049] In a tenth aspect, an embodiment of the present invention further provides a communication device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, the steps of the transmission switching method described in any one of the first to fourth aspects of the embodiments of the present invention are implemented.

[0050] For the transmission switching method, device, communication device, and storage medium provided by the embodiments of the present invention, a first network device sends a first message to a second network device, where the first message is used to request a handover; the first message includes model-related information and model data segment-related information; the first network device receives a second message sent by the second network device, where the second message indicates confirmation of the handover request; the second message includes information related to the processing of model data segments. In the technical solution of the embodiments of the present invention, by carrying model-related information and model data segment-related information in the handover request from the first network device (such as the source base station) to the second network device (such as the target base station), it is convenient for the second network device to decide whether to adopt continued transmission or retransmission for the subsequent transmission strategy of model data segments, and in the case of determining continued transmission, the starting position of the model data segments to be continuously transmitted can be determined, and information related to the processing of model data segments (such as indicating continued transmission or retransmission) is carried in the handover request confirmation message, thereby realizing the selection of an appropriate transmission method based on model-related information in the cell handover scenario, and greatly reducing the retransmission overhead on the premise of ensuring the integrity of model data after handover. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a flowchart of the transmission switching method according to an embodiment of the present invention Figure 1 ;

[0052] Figure 2 is a flowchart of the transmission switching method according to an embodiment of the present invention Figure 2 ;

[0053] Figure 3 is a flowchart of the transmission switching method according to an embodiment of the present invention Figure 3 ;

[0054] Figure 4 is a flowchart of the transmission switching method according to an embodiment of the present invention Figure 4 ;

[0055] Figure 5 is an interaction flowchart of the transmission switching method according to an embodiment of the present inventionFigure 1 ;

[0056] Figure 6 Schematic diagram of the interaction process of the transmission switching method according to an embodiment of the present invention Figure 2 ;

[0057] Figure 7 Schematic diagram of the composition structure of the transmission switching device according to an embodiment of the present invention Figure 1 ;

[0058] Figure 8 Schematic diagram of the composition structure of the transmission switching device according to an embodiment of the present invention Figure 2 ;

[0059] Figure 9 Schematic diagram of the composition structure of the transmission switching device according to an embodiment of the present invention Figure 3 ;

[0060] Figure 10 Schematic diagram of the composition structure of the transmission switching device according to an embodiment of the present invention Figure 4 ;

[0061] Figure 11 Schematic diagram of the hardware composition structure of the communication device according to an embodiment of the present invention. Detailed implementation manners

[0062] The following further describes the present invention in detail with reference to the accompanying drawings and specific embodiments.

[0063] The technical solution of the embodiment of the present invention can be applied to various communication systems, such as: Global System of Mobile communication (GSM) system, Long Term Evolution (LTE) system or 5G system, etc. Optionally, the 5G system or 5G network can also be referred to as a New Radio (NR) system or NR network.

[0064] Exemplarily, the communication system to which the embodiment of the present invention is applied may include a network device and a terminal device (which may also be referred to as a terminal, a communication terminal, etc.); the network device may be a device that communicates with the terminal device. Among them, the network device can provide communication coverage within a certain area range and can communicate with terminals located in this area. Optionally, the network device can be a base station in each communication system, such as an evolved Node B (eNB) in the LTE system, or a base station (gNB) in the 5G system or NR system.

[0065] It should be understood that in the embodiments of the present application, a device with communication functions in a network / system can be referred to as a communication device. The communication device can include a network device and a terminal with communication functions. The network device and the terminal device can be the specific devices described above, which will not be elaborated here; the communication device can also include other devices in the communication system, such as other network entities like a network controller, a mobility management entity, etc., which are not limited in the embodiments of the present invention.

[0066] It should be understood that the terms "system" and "network" are often used interchangeably in this article. The term "and / or" in this article is merely a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after.

[0067] The terms "first", "second", etc. in the description and claims of the present application are used to distinguish similar objects and do not necessarily need to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0068] The embodiments of the present invention provide a transmission switching method. Figure 1 Schematic flow of the transmission switching method according to the embodiments of the present invention Figure 1 ; as Figure 1 shown, the method includes:

[0069] Step 101: A first network device sends a first message to a second network device, where the first message is used to request a handover; the first message includes model-related information and model data segment-related information;

[0070] Step 102: The first network device receives a second message sent by the second network device, where the second message indicates confirmation of the handover request; the second message includes information related to the processing of model data segments.

[0071] In this embodiment, the first network device is the access network device to which the terminal is connected before handover. In other alternative embodiments, the first network device may also be referred to as the first base station or the source base station, etc. Correspondingly, the second network device is the access network device to which the terminal is connected after handover. In other alternative embodiments, the second network device may also be referred to as the second base station or the target base station, etc. This embodiment does not limit the names of the first network device and the second network device.

[0072] Each embodiment of the present invention is applicable to the handover process during the transmission of model data. And in this embodiment, the AI model data (such as the data of the first model) or the model data segments corresponding to the AI model (such as the first model) are stored in segments in the network device (such as the base station). It should be noted that the AI model data (such as the data of the first model) has been divided into several segments in the network device (base station side), and this segmented data becomes the model data segments. Before and after handover, at the network devices (such as the first network device and the second network device), the same segmentation strategy is adopted, that is, for the same AI model (such as the first model), the first network device and the second network device adopt the same segmentation strategy to obtain the same multiple model data segments. It can be understood that the number of model data segments for the same AI model obtained at the first network device and the second network device is the same, and the content of each corresponding model data segment is the same, etc. For example, for the same AI model data, both the first network device and the second network device have 20 model data segments, and the content included in the i-th model data segment in the first network device and the second network device is the same, where i is a positive integer less than or equal to 20.

[0073] In this embodiment, before step 101 is executed, the first network device (such as the source base station) and the terminal are already in a connected state, that is, the terminal has established a Radio Resource Control (RRC) connection with the first network device, and signaling or data can be exchanged between the terminal and the first network device.

[0074] In this embodiment, the first network device can make a handover decision based on the measurement report reported by the terminal. After determining the second network device (such as the target base station) to which the terminal is to be connected, the first network device (such as the source base station) sends a first message to the second network device (such as the target base station). In other alternative embodiments, the first message may be a HandoverRequest message to request the second network device (such as the target base station) to allocate resources for the terminal in the target cell and trigger the establishment of an X2 logical link between the first network device (such as the source base station) and the second network device (such as the target base station). This X2 logical link is used to forward the user data and related signaling cached by the first network device (such as the source base station).

[0075] In this embodiment, the first message further carries model-related information and model data segment-related information. The model-related information is used for the second network device to decide whether to retransmit or continue transmitting the model data segments; the model data segment-related information is used for the second network device to determine the starting position of the continued transmission of the model data segments when deciding to continue transmitting the model data segments.

[0076] In some alternative embodiments of the present invention, the model-related information includes second information, and the second information represents the characteristics of the first model transmitted to the terminal; the model data segment-related information includes a first value and a first identifier, the first value represents the number of model data segments that have been transmitted, and the first identifier represents the existence or non-existence of a model data segment being transmitted.

[0077] In some alternative embodiments, the second information is information associated with the model type and / or the model training method, or the second information is a model version identifier associated with the model parameters.

[0078] In this embodiment, the characteristics of the first model will affect the decision of the target base station to continue transmitting or retransmitting the model data segments, that is, it will affect whether the source base station performs a handover continuation transmission process or a handover retransmission process. The distinction between these two processes is mainly considered in combination with the specific use cases of the AI model (such as the first model). For physical layer use cases strongly related to channel scenarios such as CSI compression feedback, beam management, and positioning enhancement, the model parameters may have different values on different base stations.

[0079] For the first implementation manner, different requirements for transmission methods can be configured or determined for different model types and / or model training methods, as shown in Table 1 for example.

[0080] Table 1

[0081] Pre-trained model Re-trained model Non-transfer learning model Model transmission Switching resume transmission Switching retransmission Switching resume transmission Joint training Switching resume transmission No transmission process Switching resume transmission Separate training No transmission process No transmission process No transmission process

[0082] As shown in Table 1, the model training method can include three ways: 1. First, train the complete model on the base station side and then send it to the terminal, which corresponds to model transmission in Table 1; 2. Deploy the initial model on both the base station and terminal sides, and then enable the models deployed on both sides to interact with information such as gradients to start joint training, which corresponds to joint training in Table 1; 3. Deploy the model on both the base station and terminal sides, but do not perform model data interaction during the training of the models on both sides, which corresponds to separate training in Table 1. And for the model itself, it can be classified into a pre-trained model (transfer learning), a re-trained model (transfer learning), and a model directly obtained without transfer learning, that is, a non-transfer learning model. Among them, the pre-trained model and the non-transfer learning model are generally trained with multi-scenario data and have strong scenario generalization. Therefore, the models used between different base stations are likely to be unified, so switching and resuming transmission can be performed; for the re-trained model, the joint training and separate training schemes do not involve the transmission of the trained model, so there is no switching problem, while the joint training scheme needs to transmit the re-trained model. Considering that the re-training process is a "specialized" model for the scenario where the current base station is located, switching and re-transmission are required.

[0083] Based on this, the second network device can determine whether to perform switching and resuming transmission or switching and re-transmission by querying Table 1 based on the information associated with the model type and / or the model training method.

[0084] As a second implementation manner, the second information is a model version identifier associated with model parameters. The network device can identify the model version according to the actual situation of the AI model parameters (parameters of the first model). For example, a mapping algorithm (such as the HI algorithm, etc.) can be used to map the AI model parameters to a unique model version identifier, as shown in Table 2. The first network device carries the obtained model version identifier in the first message and sends it to the second network device. The second network device determines whether to perform switching and resuming transmission or switching and re-transmission based on the model version identifier. When the model version identifier in the first message is inconsistent with the model version identifier obtained by the second network device through the mapping algorithm, it is determined to perform switching and re-transmission; when the model version identifier in the first message is consistent with the model version identifier obtained by the second network device through the mapping algorithm, it is determined to perform switching and resuming transmission.

[0085] Table 2

[0086] Model task ID Model version identifier

[0087] In this embodiment, on the one hand, after receiving the first message, the second network device (such as the target base station) performs an admission judgment. If interruptible handover is allowed, it will interrupt the allocation of radio resources including temporary identifiers, etc. in the target cell and send a second message to the first network device (such as the source base station). The second message can be a Handover Request Acknowledge message to indicate that the handover preparation is successful; in addition, the establishment of the X2 logical channel between the base stations is completed. On the other hand, after determining handover retransmission or handover continuation based on the information carried in the first message, the second network device can also carry information related to the processing of the model data segments through the second message. The information related to the processing of the model data segments is used to indicate or trigger the first network device to execute the handover continuation process or the handover retransmission process.

[0088] In some alternative embodiments, the second message at least includes third information, and the third information indicates whether to discard or not discard the already transmitted model data segments.

[0089] In this embodiment, the third information indicates whether to discard or not discard the already transmitted model data segments, and the first network device can send a model discard notice to the terminal based on the third information to instruct the terminal to discard the already transmitted model data segments. Among them, when the third information indicates discarding the already transmitted model data segments, the first network device sends a model discard notice to the terminal.

[0090] In this embodiment, the first message may further include a first value and a first identifier. The first value represents the number of already transmitted model data segments, and the first identifier represents the existence or non - existence of a model data segment being transmitted.

[0091] Considering that the size of the model data to be transmitted is generally larger than the current signaling or data size, it is very necessary to transmit the model data in segments. There is already a mechanism for segmenting signaling in the current control plane signaling, and this mechanism can be reused when segmenting the model data transmission. Since the current segment transmission of model data through the control plane signaling does not consider the scenario of subsequent transmission during handover, signaling supplementation is required. For the possible subsequent transmission situation during handover, the first network device (such as the source base station) needs to inform the second network device (such as the target base station) of the number of segments that have been transmitted (i.e., the first value) and whether there are still model data segments being transmitted (i.e., the first identifier). Since the current model data segments are transmitted based on RRC segmentation (RRC Segmentation), the first network device (such as the source base station) can determine the transmitted model data segments according to the sequence number (SN) of the model data segments sent to the terminal, and the first identifier responsible for indicating whether there are still model data segments being transmitted will affect whether the starting point of the subsequent transmission number of the second network device (such as the target base station) continues from SN + 1 of the transmitted ones or from SN + 2 of the transmitted ones. If the first identifier indicates that there are model data segments being transmitted or model data segments that have not been transmitted completely, the first network device (such as the source base station) needs to wait until it receives an acknowledgment (ACK) message after the transmission is completed and then inform the second network device (such as the target base station) that the model data segments being transmitted have been transmitted completely.

[0092] Based on this, in some alternative embodiments of the present invention, the method further includes: the first network device sends first information to the second network device, and the first information represents the download status of the model data.

[0093] In this embodiment, if the first identifier indicates that there are model data segments being transmitted, it means that when or after the first network device sends the first message to the second network device, the first network device will send model data segments to the terminal. It can be understood that before the first network device sends the first message to the second network device, the first network device initiates the model data transmission to the terminal and sends model data segments to the terminal, for example, denoted as model data segment 1. After the first network device sends the first message to the second network device and before the terminal accesses the second network device, the first network device continues to send model data segments to the terminal, for example, denoted as model data segment 2. After the terminal receives the model data segment and sends an ACK to the first network device, the first network device sends the first information to the second network device, and the first information represents the download status of the model data, and the download status of the model data indicates that the model data segments being transmitted indicated by the first identifier have been transmitted completely.

[0094] In other alternative embodiments, if the first identifier indicates that there is no model data segment being transmitted or no incompletely transmitted model data segment, the first network device will not send the first information to the second network device.

[0095] Based on the above embodiments, an embodiment of the present invention further provides a transmission handover method. Figure 2 The flowchart of the transmission handover method according to the embodiment of the present invention Figure 2 ; as Figure 2 shown, the method includes:

[0096] Step 201: The second network device receives a first message sent by the first network device, where the first message is used to request a handover; the first message includes model-related information and model data segment-related information;

[0097] Step 202: The second network device sends a second message to the first network device, where the second message indicates confirmation of the handover request; the second message includes information related to the processing of model data segments.

[0098] In this embodiment, the first message may be a Handover Request message, which is used to request the second network device (such as the target base station) to allocate resources for the terminal in the target cell and trigger the establishment of an X2 logical link between the first network device (such as the source base station) and the second network device (such as the target base station). This X2 logical link is used to forward the user data and related signaling cached by the first network device (such as the source base station).

[0099] In this embodiment, the first message also carries model-related information and model data segment-related information. Among them, the model-related information is used for the second network device to decide whether to retransmit or continue transmitting the model data segment; the model data segment-related information is used for the second network device to determine the starting position of the model data segment to be continued when deciding to continue transmitting the model data segment.

[0100] In some alternative embodiments of the present invention, the model-related information includes second information, where the second information represents the characteristics of the first model transmitted to the terminal; the model data segment-related information includes a first value and a first identifier, where the first value represents the number of model data segments that have been transmitted, and the first identifier represents the existence or non-existence of a model data segment being transmitted.

[0101] In some alternative embodiments, the second information is information associated with the model type and / or the training method of the model, or the second information is a model version identifier associated with the model parameters.

[0102] In this embodiment, the features of the first model affect the target base station's decision on whether to resume or retransmit the segmented model data, that is, they affect whether the source base station performs a handover resume process or a handover retransmission process. The distinction between these two processes is mainly considered in combination with the specific use cases of the AI model (such as the first model). For physical layer use cases strongly related to channel scenarios such as CSI compression feedback, beam management, and positioning enhancement, the model parameters may have different values on different base stations.

[0103] For the first implementation method, different requirements for transmission methods can be configured or determined according to different model types and / or model training methods, as shown in Table 1 for example.

[0104] As shown in Table 1 above, the model training method can include three methods: 1. First, train the complete model on the base station side and then download it to the terminal, which corresponds to model transmission in Table 1; 2. Deploy the initial model on both the base station and terminal sides, and then enable the models deployed on both sides to interact with information such as gradients to start joint training, which corresponds to joint training in Table 1; 3. Deploy the model on both the base station and terminal sides, but do not perform model data interaction during the training of the models on both sides, which corresponds to separate training in Table 1. And for the model itself, it can be classified into a pre-trained model (transfer learning), a re-trained model (transfer learning), and a model directly obtained without transfer learning, that is, a non-transfer learning model. Among them, the pre-trained model and the non-transfer learning model are generally trained with multi-scenario data and have strong scenario generalization ability. Therefore, the models used between different base stations are likely to be unified, so handover resume can be performed; for the re-trained model, the joint training and separate training schemes do not involve the transmission of the trained model, so there is no handover problem, while the joint training scheme requires the transmission of the re-trained model. Considering that the re-training process is a "specialized" model for the current base station's scenario, handover retransmission is required.

[0105] Based on this, the second network device can determine whether to perform handover resume or handover retransmission by querying Table 1 based on the information associated with the model type and / or the model training method.

[0106] As a second implementation, the second information is a model version identifier associated with model parameters. The network device can identify the model version according to the actual situation of the AI model parameters (parameters of the first model). For example, a mapping algorithm (such as the HI algorithm) can be used to map the AI model parameters to a unique model version identifier, as shown in Table 2 above. The first network device carries the obtained model version identifier in the first message and sends it to the second network device. The second network device determines whether to use switch-based resume transmission or switch-based retransmission based on the model version identifier. When the model version identifier in the first message is inconsistent with the model version identifier obtained by the second network device through the mapping algorithm, it is determined to use switch-based retransmission; when the model version identifier in the first message is consistent with the model version identifier obtained by the second network device through the mapping algorithm, it is determined to use switch-based resume transmission.

[0107] In this embodiment, on the one hand, after receiving the first message, the second network device (such as the target base station) performs an admission judgment. If it allows an interrupted handover, it will allocate radio resources including a temporary identifier, etc. in the target cell interruption and send a second message to the first network device (such as the source base station). The second message can be a Handover Request Acknowledge message to indicate that the handover preparation is successful; in addition, the establishment of the X2 logical channel between the base stations is completed. On the other hand, after the second network device determines switch-based retransmission or switch-based resume transmission based on the information carried in the first message, it can also carry information related to the processing of the model data segments in the second message. The information related to the processing of the model data segments is used to indicate or trigger the first network device to execute the switch-based resume transmission process or the switch-based retransmission process.

[0108] In some alternative embodiments, the second message includes at least third information, and the third information indicates whether to discard or not discard the already transmitted model data segments.

[0109] In this embodiment, the third information indicates whether to discard or not discard the already transmitted model data segments. After the second network device determines switch-based retransmission based on the information carried in the first message, the determined third information can indicate discarding the already transmitted model data segments; after the second network device determines switch-based resume transmission based on the information carried in the first message, the determined third information can indicate not discarding the already transmitted model data segments. Then the first network device can send a model discard notification to the terminal based on the third information to instruct the terminal to discard the already transmitted model data segments. Among them, in the case where the third information indicates discarding the already transmitted model data segments, the first network device sends a model discard notification to the terminal.

[0110] In this embodiment, the first message may further include a first value and a first identifier, where the first value represents the number of segmented model data that has been transmitted, and the first identifier represents the existence or non - existence of segmented model data being transmitted.

[0111] Considering that the size of the model data to be transmitted is generally higher than the current signaling or data size, it is very necessary to transmit the model data in segments. There is already a mechanism for segmenting signaling in the current control plane signaling, and this mechanism can be reused when segmenting the model data transmission. Since the current segment - by - segment transmission of model data through the control plane signaling does not consider the scenario of subsequent transmission during handover, signaling supplementation is required. For the possible subsequent transmission situation during handover, the first network device (such as the source base station) needs to inform the second network device (such as the target base station) of the number of segments that have been transmitted (i.e., the first value) and whether there is still segmented model data being transmitted (i.e., the first identifier). Since the current segmented model data is transmitted based on RRC segmentation, the first network device (such as the source base station) can determine the transmitted segmented model data according to the sequence number (SN) of the segmented model data sent to the terminal. The first identifier responsible for indicating whether there is still segmented model data being transmitted will affect whether the starting point of the subsequent transmission number of the second network device (such as the target base station) continues from SN + 1 of the transmitted data or from SN + 2 of the transmitted data. If the first identifier indicates the existence of segmented model data being transmitted or untransmitted segmented model data, the first network device (such as the source base station) needs to wait for the confirmation (ACK) message after the transmission is completed and then inform the second network device (such as the target base station) that the segmented model data being transmitted has been completed.

[0112] Based on this, in some alternative embodiments of the present invention, the method further includes: the second network device receives the first information sent by the first network device, and the first information represents the download status of the model data.

[0113] In this embodiment, if the first identifier indicates that there is a segmented model data being transmitted, it means that when or after the first network device sends a first message to the second network device, the first network device will send the segmented model data to the terminal. It can be understood that before the first network device sends the first message to the second network device, the first network device initiates the model data transmission to the terminal and sends the segmented model data to the terminal, for example, denoted as segmented model data 1. After the first network device sends the first message to the second network device and before the terminal accesses the second network device, the first network device continues to send the segmented model data to the terminal, for example, denoted as segmented model data 2. After the terminal receives the segmented model data and sends an ACK to the first network device, the first network device sends a first information to the second network device, and the first information indicates the download status of the model data, and the download status of the model data indicates that the segmented model data being transmitted indicated by the first indication has been transmitted.

[0114] In some alternative embodiments of the present invention, the method further includes: the second network device determines a first transmission strategy, and the first transmission strategy indicates that after the terminal switches from the first network device to access the second network device, re-transmit the segmented model data of the first model or continue to transmit the segmented model data of the first model.

[0115] In this embodiment, the first transmission strategy is to determine retransmission or continued transmission after switching based on the information carried in the first message.

[0116] In some alternative embodiments, the method further includes: the second network device re-transmits the segmented model data of the first model to the terminal or continues to transmit the segmented model data of the first model.

[0117] In some alternative embodiments, a plurality of segmented model data corresponding to the first model are pre-stored in the second network device.

[0118] Embodiments of the present invention are applicable to the handover process in the model data transmission process. In this embodiment, AI model data (such as data of the first model) or model data corresponding to an AI model (such as the first model) is stored in segments in a network device (such as a base station). It should be noted that the AI model data (such as data of the first model) has been divided into several segments at the network device (base station side), and this segmented data becomes model data segments. At the network devices before and after handover (such as the first network device and the second network device), the same segmentation strategy is adopted, that is, for the same AI model (such as the first model), the first network device and the second network device adopt the same segmentation strategy to obtain the same multiple model data segments. It can be understood that the number of model data segments for the same AI model obtained at the first network device and the second network device is the same, and the content of each corresponding model data segment is the same, etc. For example, for the same AI model data, both the first network device and the second network device have 20 model data segments, and the content included in the i-th model data segment in the first network device and the second network device is the same, where i is a positive integer less than or equal to 20.

[0119] In some alternative embodiments, the continuing to transmit the model data segments of the first model includes: the second network device determines the identifier of the model data segment to be continued to be transmitted according to the first value and the first identifier, or determines the identifier of the model data segment to be continued to be transmitted according to the first identifier and the first information, and continues to transmit the model data segments of the first model to the terminal starting from the identifier of the model data segment to be continued to be transmitted.

[0120] In this embodiment, as an implementation manner, the second network device may determine the identifier of the model data segment to be continuously transmitted based on the first value and the first identifier. If the first identifier indicates that there is no model data segment being transmitted, the second network device may determine the identifier of the model data segment to be continuously transmitted based on the first value (such as the SN of the model data segment), that is, use SN+1 as the identifier of the model data segment to be continuously transmitted. If the first identifier indicates that there is a model data segment being transmitted, the second network device may determine the identifier of the model data segment to be continuously transmitted based on the first value (such as the SN of the model data segment) and the first identifier. For example, use SN+2 as the identifier of the model data segment to be continuously transmitted (assuming the number of model data segments being transmitted is 1). As another implementation manner, the second network device may determine the identifier of the model data segment to be continuously transmitted according to the first identifier and the first information. Since the first information represents the download status of the model data, specifically indicating that the model data segment being transmitted indicated by the first identifier has been transmitted, the second network device may determine the identifier of the model data segment to be continuously transmitted based on the first value (such as the SN of the model data segment) and the download status of the model data represented by the first information. For example, use SN+2 as the identifier of the model data segment to be continuously transmitted (assuming the number of model data segments being transmitted is 1).

[0121] An embodiment of the present invention also provides a transmission switching method. Figure 3 The flowchart of the transmission switching method according to the embodiment of the present invention Figure 3 ; as Figure 3 shown, the method includes:

[0122] Step 301: The first network device initiates the data transmission of the first model and sends the model data segments of the first model to the terminal;

[0123] Step 302: When the first network device makes a handover decision and determines that the terminal is connected to the second network device, continue to transmit the model data segments of the first model that have not been transmitted yet to the terminal, or send the fourth information to the terminal, where the fourth information is used to notify to discard the model data segments of the first model that have been transmitted.

[0124] In some alternative embodiments of the present invention, a plurality of model data segments corresponding to the first model are pre-stored in the first network device.

[0125] Embodiments of the present invention are applicable to the handover process in the model data transmission process. In this embodiment, AI model data (such as the data of the first model) or model data segments corresponding to an AI model (such as the first model) are stored in segments in a network device (such as a base station). It should be noted that the AI model data (such as the data of the first model) has been divided into several segments in the network device (base station side), and this segmented data becomes model data segments. Before and after handover, at network devices (such as the first network device and the second network device), the same segmentation strategy is adopted, that is, for the same AI model (such as the first model), the first network device and the second network device adopt the same segmentation strategy to obtain the same multiple model data segments. It can be understood that the number of model data segments for the same AI model obtained at the first network device and the second network device is the same, and the content of each corresponding model data segment is the same, etc. For example, for the same AI model data, both the first network device and the second network device have 20 model data segments, and the content included in the i-th model data segment in the first network device and the second network device is the same, where i is a positive integer less than or equal to 20.

[0126] In this embodiment, the first network device initiates the data transmission of the first model and sends the model data segments of the first model to the terminal. Among them, the sent model data segments can be the model data segments sent in ascending order according to a preset identifier. For example, model data segment 1, model data segment 2... are sent in sequence, and so on.

[0127] In some alternative embodiments, the sending the model data segments of the first model to the terminal includes: the first network device sending the model data segments of the first model to the terminal through RRC signaling; and / or, the continuing to transmit the model data segments of the first model that have not been completely transmitted to the terminal includes: the first network device continuing to transmit the model data segments of the first model that have not been completely transmitted to the terminal through RRC signaling.

[0128] In this embodiment, before step 301 is executed, the first network device (such as the source base station) and the terminal are already in a connected state, that is, the terminal has established an RRC connection with the first network device, and signaling or data can be exchanged between the terminal and the first network device. In this embodiment, the first network device transmits the model data segments of the first model to the terminal through control plane signaling (such as RRC signaling).

[0129] In some alternative embodiments of the present invention, when the first network device makes a handover decision to determine that the terminal accesses the second network device, the method further includes: the first network device sending a first message to the second network device, where the first message is used to request a handover; the first message includes model-related information and model data segment-related information; the first network device receiving a second message sent by the second network device, where the second message indicates confirmation of the handover request; the second message includes information related to the processing of model data segments.

[0130] In some alternative embodiments, the model-related information includes second information, where the second information represents the characteristics of the first model transmitted to the terminal; the model data segment-related information includes a first value and a first identifier, where the first value represents the number of model data segments that have been transmitted, and the first identifier represents the existence or non-existence of a model data segment being transmitted; and / or, the second message at least includes third information, where the third information represents discarding or not discarding the transmitted model data segments.

[0131] In some alternative embodiments, the second information is information associated with the model type and / or the training method of the model, or the second information is a model version identifier associated with the model parameters.

[0132] In some alternative embodiments of the present invention, continuing to transmit the model data segments of the first model that have not been transmitted completely to the terminal includes: when the first identifier represents the existence of a model segment data being transmitted, the first network device continuing to transmit the model data segments of the first model that have not been transmitted completely to the terminal.

[0133] In some alternative embodiments of the present invention, sending fourth information to the terminal includes: when the third information represents discarding the transmitted model data segments, sending fourth information to the terminal.

[0134] In this embodiment, when the first network device receives the second message sent by the second network device, and the third information included in the second message represents discarding the transmitted model data segments, the first network device sends the fourth information, where the fourth information is used to notify discarding the model data segments of the first model that have been transmitted.

[0135] Based on the above embodiments, an embodiment of the present invention further provides a transmission handover method. Figure 4 Schematic flow of the transmission handover method according to the embodiment of the present invention Figure 4 ; as Figure 4 shown, the method includes:

[0136] Step 401: The terminal receives the segmented model data of the first model transmitted by the first network device;

[0137] Step 402: When the terminal switches from the first network device to the second network device, the terminal receives the segmented model data of the first model that has not been completely transmitted by the first network device, or receives the fourth information sent by the first network device, where the fourth information is used to notify discarding the segmented model data of the first model that has been transmitted.

[0138] In this embodiment, the first network device initiates the data transmission of the first model and sends the segmented model data of the first model to the terminal. Among them, the sent segmented model data can be the segmented model data sent in ascending order according to a preset identifier. For example, the segmented model data 1, segmented model data 2... are sent in sequence, and so on.

[0139] In some optional embodiments, the terminal receiving the segmented model data of the first model transmitted by the first network device includes: the terminal receiving the segmented model data of the first model transmitted by the first network device through RRC signaling; and / or, the receiving the segmented model data of the first model that has not been completely transmitted by the first network device includes: the terminal receiving the segmented model data of the first model that has not been completely transmitted by the first network device through RRC signaling.

[0140] In this embodiment, before step 401 is executed, the first network device (such as the source base station) and the terminal are already in a connected state, that is, the terminal has established an RRC connection with the first network device, and signaling or data interaction can be performed between the terminal and the first network device. In this embodiment, the first network device transmits the segmented model data of the first model to the terminal through control plane signaling (such as RRC signaling).

[0141] In some optional embodiments, when the terminal receives the fourth information sent by the first network device, the method further includes: the terminal discarding the segmented model data of the first model that has been received.

[0142] In this embodiment, after the second network device determines the first transmission policy (such as handover retransmission or handover continuation transmission), it sends a second message carrying information related to the processing of the segmented model data to the first network device. The first network device performs a handover continuation transmission process or a handover retransmission process based on the second message. In the case where the first network device performs a handover retransmission process, the first network device sends the fourth information to the terminal to notify discarding the segmented model data of the first model that has been transmitted; then the terminal discards the segmented model data of the first model that has been received based on the fourth information. Among them, the terminal discarding the segmented model data of the first model that has been received may include the first segmented model data and may also include the second segmented model data.

[0143] In some alternative embodiments, the method further includes: the terminal receives the segmented model data of the first model retransmitted by the second network device.

[0144] In some other alternative embodiments, the method further includes: the terminal receives the segmented model data of the first model continuously transmitted by the second network device.

[0145] In this embodiment, after the terminal accesses the second network device, it can receive the segmented model data of the first model from the second network device according to the first transmission policy (such as handover retransmission or handover continuation transmission) determined by the second network device. One implementation manner is that the first transmission policy is a handover retransmission policy, then the terminal discards the segmented model data of the first model that has been received and receives the segmented model data of the first model retransmitted by the second network device, specifically starting to receive the segmented model data of the first model from the first segmented model data of the first model. Another implementation manner is that the first transmission policy is a handover continuation transmission policy, then the terminal receives the segmented model data of the first model continuously transmitted by the second network device.

[0146] The technical solutions of the embodiments of the present invention will be described in detail below with specific examples. In the following examples, the first network device is taken as the source base station and the second network device is taken as the target base station for illustration.

[0147] Figure 5 Schematic diagram of the interaction process of the transmission handover method for the embodiments of the present invention Figure 1 ; as Figure 5 shown, the method includes:

[0148] Step 501: After the user equipment (UE) establishes an RRC connection with the source base station, the source base station initiates the transmission of the segmented model data of the first model to the terminal.

[0149] Step 502: The source base station sequentially transmits the model data segments by using RRC signaling; for example, the source base station sends model data segment 1 and model data segment 2 to the UE.

[0150] Step 503: The UE reports a measurement report to the source base station to feedback the current channel transmission quality.

[0151] Step 504: The source base station makes a handover decision based on the measurement report; after determining to perform a handover, the target base station is determined.

[0152] Step 505: The source base station sends a handover request message to the target base station. The handover request message includes second information, a first value, and a first identifier. The second information represents the characteristics of the first model, the first value represents the number of model data segments that have been transmitted, and the first identifier represents the existence or non-existence of a model data segment being transmitted.

[0153] Here, the handover request message is equivalent to the first message in the above embodiment.

[0154] Step 506: The target base station feeds back a handover request confirmation message to the source base station. The handover request confirmation message includes third information, and the third information represents not discarding the model data segments that have been transmitted.

[0155] Here, the handover request confirmation message is equivalent to the second message in the above embodiment.

[0156] Step 507: The source base station continues to transmit the unfinished segments to the UE by using RRC signaling; for example, continues to transmit model data segment 3 to the UE.

[0157] Step 508: The terminal and the source base station perform RRC reconfiguration.

[0158] Step 509: After the unfinished segments are completely transmitted, the source base station sends first information to the target base station, and the first information represents the download status of the model data.

[0159] It should be noted that steps 507 and 509 are optional steps.

[0160] Step 510: The terminal accesses the target base station.

[0161] Step 511: The target base station starts to continue transmitting the model data segments of the first model according to the first value and the first identifier received in step 505, or according to the first identifier and the first information.

[0162] Here, without including step 507 and step 509, the target base station determines to continue transmitting the model data segment of the first model according to the second information in step 505, and then determines, according to the first value and the first identifier, that it is necessary to continue transmitting the model data segment starting from model data segment 3. In the case of including step 507 and step 509, it is determined, according to the first identifier and the first information, that model data segment 3 has been transmitted completely, and then it is determined that it is necessary to continue transmitting the model data segment starting from model data segment 4.

[0163] Step 512: The target base station continues the model data segment by using RRC signaling.

[0164] Step 513: The transmission of the model data segment of the first model is completed.

[0165] Figure 6 Schematic diagram of the interaction process of the transmission handover method according to the embodiment of the present invention Figure 2 ; as Figure 6 shown, the method includes:

[0166] Step 601: After the UE establishes an RRC connection with the source base station, the source base station initiates the transmission of the model data segment of the first model for the terminal.

[0167] Step 602: The source base station sequentially transmits the model data segments by using RRC signaling; for example, the source base station sends model data segment 1 and model data segment 2 to the UE.

[0168] Step 603: The UE reports a measurement report to the source base station to feedback the current channel transmission quality.

[0169] Step 604: The source base station makes a handover decision according to the measurement report; after determining to perform handover, the target base station is determined.

[0170] Step 605: The source base station sends a handover request message to the target base station. The handover request message includes second information, a first value, and a first identifier. The second information represents the characteristics of the first model, the first value represents the number of the model data segments that have been transmitted, and the first identifier represents the existence or non-existence of a model data segment being transmitted.

[0171] Here, the handover request message is equivalent to the first message in the above embodiment.

[0172] Step 606: The target base station feeds back a handover request confirmation message to the source base station. The handover request confirmation message includes third information, and the third information represents discarding the model data segments that have been transmitted.

[0173] Here, the handover request confirmation message is equivalent to the second message in the above embodiment.

[0174] Step 607: The source base station sends a model discard notification to the UE, and the UE discards the segments of the received model data according to the model discard notification; for example, the UE discards the received model data segment 1 and model data segment 2.

[0175] Here, the model discard notification is equivalent to the fourth information in the above embodiment.

[0176] Step 608: The terminal and the source base station perform RRC reconfiguration.

[0177] Step 609: The terminal accesses the target base station.

[0178] Step 610: The target base station initiates the retransmission of the segments of the model data of the first model.

[0179] Step 611: The target base station uses RRC signaling to re - send the segments of the model data of the first model to the UE starting from model data segment 1.

[0180] Step 612: The transmission of the segments of the model data of the first model is completed.

[0181] It should be noted that in the above example, the source base station and the target base station pre - store the data of the first model and adopt the same segmentation strategy to obtain the same multiple model data segments. For example, the model data segments are denoted as model data segment 1, 2, … n.

[0182] Based on the above embodiments, an embodiment of the present invention further provides a transmission switching device, and the device is applied to a first network device. Figure 7 Schematic diagram of the composition structure of the transmission switching device according to the embodiment of the present invention Figure 1 ; as Figure 7 shown, the device includes a first communication unit 11, configured to send a first message to a second network device, where the first message is used to request a handover; the first message includes model - related information and model data segment - related information; and is further configured to receive a second message sent by the second network device, where the second message indicates a handover request confirmation; the second message includes information related to the processing of the model data segments.

[0183] In some alternative embodiments of the present invention, the first communication unit 11 is further configured to send a first information to the second network device, where the first information indicates the download status of the model data.

[0184] In some alternative embodiments of the present invention, the model - related information includes second information, where the second information represents the characteristics of the first model transmitted to the terminal; the model data segment - related information includes a first value and a first identifier, the first value represents the number of the transmitted model data segments, and the first identifier represents the existence or non - existence of a model data segment being transmitted.

[0185] In some alternative embodiments of the present invention, the second information is information associated with the model type and / or the training method of the model, or the second information is a model version identifier associated with the model parameters.

[0186] In some alternative embodiments of the present invention, the second message at least includes third information, and the third information indicates whether to discard or not to discard the transmitted model data segments.

[0187] In the embodiments of the present invention, the first communication unit 11 in the device can be implemented by a communication module (including: basic communication suite, operating system, communication module, standardized interface and protocol, etc.) and a transceiver antenna in practical applications.

[0188] The embodiments of the present invention further provide a transmission switching device, and the device is applied to a second network device. Figure 8 Schematic diagram of the composition structure of the transmission switching device according to the embodiments of the present invention Figure 2 ; as Figure 8 shown, the device includes a second communication unit 21, configured to receive a first message sent by a first network device, where the first message is used to request a handover; the first message includes model-related information and model data segment-related information; and is further configured to send a second message to the first network device, where the second message indicates a handover request confirmation; the second message includes information related to the processing of the model data segments.

[0189] In some alternative embodiments of the present invention, the second communication unit 21 is further configured to receive a first information sent by the first network device, where the first information indicates the download status of the model data.

[0190] In some alternative embodiments of the present invention, the model-related information includes second information, and the second information represents the characteristics of the first model transmitted to the terminal; the model data segment-related information includes a first value and a first identifier, the first value represents the number of transmitted model data segments, and the first identifier indicates the existence or non-existence of a model data segment being transmitted.

[0191] In some alternative embodiments of the present invention, the second information is information associated with the model type and / or the training method of the model, or the second information is a model version identifier associated with the model parameters.

[0192] In some alternative embodiments of the present invention, the second message at least includes third information, and the third information indicates whether to discard or not to discard the transmitted model data segments.

[0193] In some alternative embodiments of the present invention, the second communication unit 21 is further configured to re-transmit the model data segments of the first model to the terminal or continue to transmit the model data segments of the first model to the terminal.

[0194] In some alternative embodiments of the present invention, the apparatus further includes a first processing unit 22, configured to determine the identifier of the model data segment to be continuously transmitted according to the first value and the first identifier, or determine the identifier of the model data segment to be continuously transmitted according to the first identifier and the first information;

[0195] The second communication unit 21 is configured to continue to transmit the model data segments of the first model to the terminal starting from the identifier of the model data segment to be continuously transmitted.

[0196] In some alternative embodiments of the present invention, the apparatus further includes a first storage unit, configured to pre-store a plurality of model data segments corresponding to the first model.

[0197] In the embodiments of the present invention, the first processing unit 22 in the apparatus can be implemented by a central processing unit (CPU, Central Processing Unit), a digital signal processor (DSP, Digital Signal Processor), a microcontroller unit (MCU, Microcontroller Unit), or a field-programmable gate array (FPGA, Field-Programmable Gate Array) in practical applications; the second communication unit 21 in the apparatus can be implemented by a communication module (including: basic communication suite, operating system, communication module, standardized interface and protocol, etc.) and a transceiver antenna in practical applications.

[0198] The embodiments of the present invention further provide a transmission switching apparatus, which is applied to a first network device. Figure 9 Schematic diagram of the composition structure of the transmission switching apparatus according to the embodiments of the present invention Figure 3 ; as Figure 9 shown, the apparatus includes a third communication unit 31, configured to initiate the data transmission of the first model, send the model data segments of the first model to the terminal; and is further configured to, in the case of determining that the terminal accesses a second network device during the handover decision, continue to transmit the model data segments of the first model that have not been completely transmitted to the terminal, or send a fourth piece of information to the terminal, where the fourth piece of information is used to notify to discard the model data segments of the first model that have been transmitted.

[0199] In some alternative embodiments of the present invention, the third communication unit 31 is configured to send, to a terminal, model data segments of the first model via RRC signaling; and / or continue to transmit, to the terminal via RRC signaling, model data segments of the first model that have not been completely transmitted yet.

[0200] In some alternative embodiments of the present invention, when the third communication unit 31 is further configured to make a handover decision to determine that the terminal accesses a second network device, the third communication unit 31 is configured to send a first message to the second network device, where the first message is used to request a handover; the first message includes model-related information and information related to model data segments; receive a second message sent by the second network device, where the second message indicates confirmation of the handover request; and the second message includes information related to the processing of model data segments.

[0201] In some alternative embodiments of the present invention, the model-related information includes second information, where the second information represents features of the first model transmitted to the terminal; the information related to model data segments includes a first value and a first identifier, where the first value represents the number of model data segments that have been transmitted, and the first identifier represents the existence or non-existence of a model data segment that is being transmitted; and / or

[0202] The second message at least includes third information, where the third information represents discarding or not discarding the transmitted model data segments.

[0203] In some alternative embodiments of the present invention, the second information is information associated with a model type and / or a model training method, or the second information is a model version identifier associated with model parameters.

[0204] In some alternative embodiments of the present invention, when the first identifier represents the existence of a model segment data that is being transmitted, the third communication unit 31 is configured to continue to transmit, to the terminal, model data segments of the first model that have not been completely transmitted yet.

[0205] In some alternative embodiments of the present invention, when the third information represents discarding the transmitted model data segments, the third communication unit 31 is configured to send fourth information to the terminal.

[0206] In some alternative embodiments of the present invention, the device further includes a second storage unit configured to pre-store a plurality of model data segments corresponding to the first model.

[0207] In an embodiment of the present invention, the third communication unit 31 in the device can be implemented by a communication module (including: a basic communication suite, an operating system, a communication module, a standardized interface, and a protocol, etc.) and a transceiver antenna in practical applications.

[0208] An embodiment of the present invention further provides a transmission switching device, and the device is applied to a terminal. Figure 10 Schematic diagram of the composition structure of the transmission switching device according to the embodiment of the present invention Figure 4 ; as Figure 10 As shown, the device includes a fourth communication unit 41, configured to receive segmented model data of a first model transmitted by a first network device; and is further configured to, in the case of switching access from the first network device to a second network device, receive the segmented model data of the first model that has not been completely transmitted by the first network device, or receive a fourth piece of information sent by the first network device, where the fourth piece of information is used to notify discarding the segmented model data of the first model that has been transmitted.

[0209] In some optional embodiments of the present invention, the fourth communication unit 41 is configured to receive segmented model data of a first model transmitted by a first network device through an RRC signaling; and / or receive the segmented model data of the first model that has not been completely transmitted by the first network device through an RRC signaling.

[0210] In some optional embodiments of the present invention, the device further includes a second processing unit 42, configured to discard the segmented model data of the first model that has been received in the case where the fourth communication unit receives the fourth piece of information sent by the first network device.

[0211] In some optional embodiments of the present invention, the fourth communication unit 41 is further configured to receive the segmented model data of the first model retransmitted by a second network device.

[0212] In some optional embodiments of the present invention, the fourth communication unit 41 is further configured to receive the segmented model data of the first model continuously transmitted by the second network device.

[0213] In some optional embodiments of the present invention, the first network device and the second network device pre-store multiple segmented model data corresponding to the first model.

[0214] In the embodiment of the present invention, the second processing unit 42 in the device can be implemented by a CPU, a DSP, an MCU, or an FPGA in practical applications; the fourth communication unit 41 in the device can be implemented by a communication module (including: a basic communication suite, an operating system, a communication module, a standardized interface, and a protocol, etc.) and a transceiver antenna in practical applications.

[0215] It should be noted that when the transmission switching device provided in the above embodiments performs transmission switching, only the division of the above program modules is used for illustration. In actual applications, the above processing can be allocated to different program modules according to needs, that is, the internal structure of the device is divided into different program modules to complete all or part of the above-described processing. In addition, the transmission switching device provided in the above embodiments and the embodiments of the transmission switching method belong to the same concept. For the specific implementation process, please refer to the method embodiments and will not be elaborated here.

[0216] An embodiment of the present invention further provides a communication device, which may specifically be a first network device, a second network device, or a terminal. Figure 11 It is a schematic diagram of the hardware composition structure of the communication device according to the embodiment of the present invention, as Figure 11 shown, the communication device includes a memory 52, a processor 51, and a computer program stored on the memory 52 and executable on the processor 51. When the processor 51 executes the program, it implements the steps of the transmission switching method applied to the first network device, the second network device, or the terminal.

[0217] Optionally, the communication device further includes at least one network interface 53. Among them, the various components in the communication device are coupled together through a bus system 54. It can be understood that the bus system 54 is used to realize the connection and communication between these components. The bus system 54 includes not only a data bus, but also a power bus, a control bus, and a status signal bus. However, for the sake of clarity, in Figure 11 all kinds of buses are labeled as the bus system 54.

[0218] It can be understood that the memory 52 can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM, Read Only Memory), a programmable read-only memory (PROM, Programmable Read-Only Memory), an erasable programmable read-only memory (EPROM, Erasable Programmable Read-Only Memory), an electrically erasable programmable read-only memory (EEPROM, Electrically Erasable Programmable Read-Only Memory), a ferromagnetic random access memory (FRAM, Ferromagnetic Random Access Memory), a flash memory (Flash Memory), a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM, Compact Disc Read-Only Memory); the magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a random access memory (RAM, Random Access Memory), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as a static random access memory (SRAM, Static Random Access Memory), a synchronous static random access memory (SSRAM, Synchronous Static Random Access Memory), a dynamic random access memory (DRAM, Dynamic Random Access Memory), a synchronous dynamic random access memory (SDRAM, Synchronous Dynamic Random Access Memory), a double data rate synchronous dynamic random access memory (DDR SDRAM, Double Data Rate Synchronous Dynamic Random Access Memory), an enhanced synchronous dynamic random access memory (ESDRAM, Enhanced Synchronous Dynamic Random Access Memory), a sync link dynamic random access memory (SLDRAM, SyncLink Dynamic Random Access Memory), and a direct rambus random access memory (DRRAM, Direct Rambus Random Access Memory).The memory 52 described in the embodiments of the present invention is intended to include, but is not limited to, these and any other suitable types of memories.

[0219] The method disclosed in the embodiments of the present invention above can be applied to the processor 51 or implemented by the processor 51. The processor 51 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the integrated logic circuit in the hardware of the processor 51 or the instructions in the form of software. The above-mentioned processor 51 may be a general-purpose processor, DSP, or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 51 can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present invention. The general-purpose processor may be a microprocessor or any conventional processor, etc. Combining the steps of the method disclosed in the embodiments of the present invention, it can be directly embodied as being executed and completed by the hardware decoding processor, or executed and completed by the combination of the hardware and software modules in the decoding processor. The software module may be located in the storage medium, which is located in the memory 52. The processor 51 reads the information in the memory 52 and combines its hardware to complete the steps of the foregoing method.

[0220] In an exemplary embodiment, the communication device may be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), FPGAs, general-purpose processors, controllers, MCUs, microprocessors, or other electronic components for performing the foregoing method.

[0221] In an exemplary embodiment, the embodiments of the present invention also provide a computer-readable storage medium, such as the memory 52 including a computer program, and the above computer program can be executed by the processor 51 of the communication device to complete the steps of the foregoing method. The computer-readable storage medium may be a FRAM, ROM, PROM, EPROM, EEPROM, Flash Memory, magnetic surface memory, optical disc, or CD-ROM, etc.; it may also be various devices including one or any combination of the above memories.

[0222] The computer-readable storage medium provided by the embodiments of the present invention stores a computer program thereon, and when the program is executed by the processor, it implements the steps of the transmission switching method of the embodiments of the present invention applied to the first network device, the second network device, or the terminal.

[0223] In the method embodiments disclosed in several method embodiments provided by this application, they can be arbitrarily combined without conflict to obtain new method embodiments.

[0224] In the features disclosed in several product embodiments provided by this application, they can be arbitrarily combined without conflict to obtain new product embodiments.

[0225] In the features disclosed in several method or device embodiments provided by this application, they can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0226] In several embodiments provided by this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed with each other can be through some interfaces. The indirect coupling or communication connection of devices or units can be electrical, mechanical, or other forms.

[0227] The units described above as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0228] In addition, in each embodiment of the present invention, each functional unit can be all integrated in a processing unit, or each unit can be separately used as a unit, or two or more units can be integrated in a unit; the above-mentioned integrated units can be implemented in the form of hardware, or can be implemented in the form of hardware plus software functional units.

[0229] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps including the above method embodiments; and the foregoing storage medium includes: various media such as mobile storage devices, ROM, RAM, magnetic disks, or optical discs that can store program codes.

[0230] Alternatively, if the above-integrated units of the present invention are implemented in the form of software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiments of the present invention essentially or the part that contributes to the prior art can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the various embodiments of the present invention. The foregoing storage medium includes: various media such as removable storage devices, ROM, RAM, magnetic disks, or optical discs that can store program codes.

[0231] As described above, the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A transmission switching method, characterized in that, The method is applied to a first network device, and the method includes: The first network device sends a first message to a second network device, where the first message is used to request a handover; the first message includes model-related information and model data segment-related information; The first network device receives a second message sent by the second network device, where the second message indicates confirmation of the handover request; the second message includes information related to the processing of model data segments.

2. The method according to claim 1, wherein The method further includes: The first network device sends first information to the second network device, where the first information indicates the download status of model data.

3. The method according to claim 1, wherein The model-related information includes second information, where the second information indicates the characteristics of a first model transmitted to a terminal; the model data segment-related information includes a first value and a first identifier, where the first value represents the number of model data segments that have been transmitted, and the first identifier represents the existence or non-existence of a model data segment being transmitted.

4. The method according to claim 3, wherein The second information is information associated with the model type and / or the training method of the model, or the second information is a model version identifier associated with model parameters.

5. The method according to claim 1, wherein The second message at least includes third information, where the third information indicates discarding or not discarding the transmitted model data segments.

6. A transmission switching method, characterized in that, The method is applied to a second network device, and the method includes: The second network device receives a first message sent by a first network device, where the first message is used to request a handover; the first message includes model-related information and model data segment-related information; The second network device sends a second message to the first network device, where the second message indicates confirmation of the handover request; the second message includes information related to the processing of model data segments.

7. The method according to claim 6, wherein The method further includes: The second network device receives first information sent by the first network device, where the first information indicates the download status of model data.

8. The method according to claim 6, wherein The model-related information includes second information, where the second information indicates the characteristics of a first model transmitted to a terminal; the model data segment-related information includes a first value and a first identifier, where the first value represents the number of model data segments that have been transmitted, and the first identifier represents the existence or non-existence of a model data segment being transmitted.

9. The method according to claim 8, wherein The method further includes: The second network device re-transmits or continues to transmit model data segments of the first model to the terminal.

10. The method according to claim 9, characterized in that, The continuing to transmit model data segments of the first model includes: The second network device determines the identifier of the model data segment to continue transmitting according to the first value and the first identifier, or determines the identifier of the model data segment to continue transmitting according to the first identifier and the first information, and continues to transmit model data segments of the first model to the terminal starting from the identifier of the model data segment to continue transmitting.

11. The method according to claim 8, wherein A plurality of model data segments corresponding to the first model are pre-stored in the second network device.

12. A transmission switching method, characterized in that, The method is applied to a first network device, and the method includes: The first network device initiates data transmission of a first model and sends model data segments of the first model to a terminal; When the first network device makes a handover decision to determine that the terminal accesses the second network device, continue to transmit to the terminal the model data segments of the first model that have not been completely transmitted, or send the fourth information to the terminal, where the fourth information is used to notify discarding the model data segments of the first model that have been transmitted.

13. The method according to claim 12, wherein The transmitting the model data segments of the first model to the terminal includes: The first network device transmits the model data segments of the first model to the terminal through radio resource control (RRC) signaling; and / or, The continuing to transmit to the terminal the model data segments of the first model that have not been completely transmitted includes: The first network device continues to transmit to the terminal the model data segments of the first model that have not been completely transmitted through RRC signaling.

14. The method according to claim 12, wherein When the first network device makes a handover decision to determine that the terminal accesses the second network device, the method further includes: The first network device sends a first message to the second network device, where the first message is used to request a handover; the first message includes model-related information and model data segment-related information; The first network device receives a second message sent by the second network device, where the second message indicates confirmation of the handover request; the second message includes information related to the processing of model data segments.

15. The method according to claim 14, characterized in that, The model-related information includes second information, where the second information represents the features of the first model transmitted to the terminal; the model data segment-related information includes a first value and a first identifier, where the first value represents the number of model data segments that have been transmitted, and the first identifier represents the existence or non-existence of a model data segment that is being transmitted; and / or, The second message at least includes third information, where the third information represents discarding or not discarding the model data segments that have been transmitted.

16. The method according to claim 15, wherein The second information is information associated with the model type and / or the training method of the model, or the second information is a model version identifier associated with model parameters.

17. The method according to claim 15, characterized in that, The continuing to transmit to the terminal the model data segments of the first model that have not been completely transmitted includes: When the first identifier represents the existence of a model segment data that is being transmitted, the first network device continues to transmit to the terminal the model data segments of the first model that have not been completely transmitted.

18. The method according to claim 15, wherein The sending the fourth information to the terminal includes: When the third information represents discarding the model data segments that have been transmitted, send the fourth information to the terminal.

19. The method according to claim 12, characterized in that, Multiple model data segments corresponding to the first model are pre-stored in the first network device.

20. A transmission switching method, characterized in that The method is applied to a terminal, and the method includes: The terminal receives the model data segments of the first model transmitted by the first network device; When the terminal is handed over from the first network device to access the second network device, receive the model data segments of the first model that have not been completely transmitted continuously transmitted by the first network device, or receive the fourth information sent by the first network device, where the fourth information is used to notify discarding the model data segments of the first model that have been transmitted.

21. The method according to claim 20, characterized in that, When the terminal receives the fourth information sent by the first network device, the method further includes: The terminal discards the model data segments of the first model that have been received.

22. The method according to claim 21, wherein The method further includes: The terminal receives the model data segments of the first model retransmitted by the second network device.

23. The method according to claim 20, wherein The method further includes: The terminal receives the model data segments of the first model continuously transmitted by the second network device.

24. The method according to claim 22 or 23, characterized in that, The first network device and the second network device pre-store a plurality of model data segments corresponding to the first model.

25. A transmission switching device, characterized in that, The device is applied to a first network device. The device includes a first communication unit configured to send a first message to a second network device. The first message is used to request a handover. The first message includes model-related information and model data segment-related information. The first communication unit is further configured to receive a second message sent by the second network device. The second message indicates confirmation of the handover request. The second message includes information related to the processing of model data segments.

26. A transmission switching device, characterized in that, The device is applied to a second network device. The device includes a second communication unit configured to receive a first message sent by a first network device. The first message is used to request a handover. The first message includes model-related information and model data segment-related information. The second communication unit is further configured to send a second message to the first network device. The second message indicates confirmation of the handover request. The second message includes information related to the processing of model data segments.

27. A transmission switching device, characterized in that, The device is applied to a first network device. The device includes a third communication unit configured to initiate data transmission of a first model and send model data segments of the first model to a terminal. The third communication unit is further configured to, when determining in a handover decision that the terminal accesses the second network device, continue to transmit the model data segments of the first model that have not been completely transmitted to the terminal, or send fourth information to the terminal. The fourth information is used to notify discarding the model data segments of the first model that have been transmitted.

28. A transmission switching device, characterized in that, The device is applied to a terminal. The device includes a fourth communication unit configured to receive model data segments of a first model transmitted by a first network device. The fourth communication unit is further configured to, when switching from the first network device to access the second network device, receive the model data segments of the first model that have not been completely transmitted continuously transmitted by the first network device, or receive fourth information sent by the first network device. The fourth information is used to notify discarding the model data segments of the first model that have been transmitted.

29. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 5; or, When the program is executed by a processor, it implements the steps of the method according to any one of claims 6 to 11; or, When the program is executed by a processor, it implements the steps of the method according to any one of claims 12 to 19; or, When the program is executed by a processor, it implements the steps of the method according to any one of claims 20 to 24.

30. A communication device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method according to any one of claims 1 to 5; or, When the processor executes the program, it implements the steps of the method according to any one of claims 6 to 11; or, When the processor executes the program, the steps of the method according to any one of claims 12 to 19 are implemented; Or, When the processor executes the program, the steps of the method according to any one of claims 20 to 24 are implemented.

Images