Communication method and apparatus

Abstract

A communication method and apparatus, which are used for implementing registration of an intelligent sensing model. The method comprises: a terminal device sends registration information of a sensing model. The registration information of the sensing model may comprise one or more of the following: a sensing model identifier, sensing model performance information, sensing model input information, or a sensing model availability condition. By means of the foregoing method, registration of the sensing model can be implemented, enabling a network device to manage the sensing model. Further, sharing of sensing models of a plurality of terminal devices can be implemented in a communication system, thereby improving the sensing performance of the communication system.

Description

A communication method and apparatus

[0001] Cross-reference to Related Applications

[0002] This application claims priority to the Chinese Patent Application No. 202410904432.9, filed on July 5, 2024, and entitled “A communication method and apparatus”, the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0003] The present application relates to the field of communication technology, and in particular to a communication method and apparatus. BACKGROUND

[0004] Wireless sensing technology is to realize the sensing of a target by receiving the echo of the sensing target, wherein the sensed target (also referred to as the sensing target) can reflect, diffract or scatter signals, etc.

[0005] With the development of artificial intelligence technology, artificial intelligence (AI) models or machine learning (ML) models are widely applied in various fields, especially in the sensing field and have begun to attract a lot of attention. The process of intelligent sensing based on AI or ML usually includes the following parts: a data collection node (terminal device or network device) collects sensing data, and trains or reasons an intelligent sensing model according to the collected sensing data.

[0006] In order to facilitate the network device to manage the AI or ML model, the model of the terminal device needs to be registered to the network device, so that after the model is registered, the network device can manage the model (such as performance monitoring, cooperative training, shared transmission, etc.). Therefore, how to realize the registration of the intelligent sensing model is a problem to be solved. SUMMARY

[0007] Embodiments of the present application provide a communication method and apparatus to realize the registration of an intelligent sensing model.

[0008] In a first aspect, the present application provides a communication method, which can be applied to a communication apparatus. The communication apparatus can be a terminal device, or can be a component (such as a processor, a chip, a chip system, a circuit or a functional module, etc.) in the terminal device. The method can include: sending registration information of a sensing model, wherein the registration information of the sensing model can include one or more of the following: a sensing model identifier, sensing model performance information, sensing model input information or sensing model available conditions.

[0009] By the method, the registration of the perception model can be implemented, so that the network device manages the perception model, and then the perception models of the plurality of terminal devices can be shared in the communication system, and the perception performance of the communication system is improved.

[0010] In a possible design, the perception model identifier can include a perception model function identifier and a perception model architecture identifier, or the perception model identifier includes the perception model function identifier, the perception model architecture identifier and a terminal device identifier. In this way, the perception model identifier can be flexibly represented by a global identifier or a local identifier.

[0011] In a possible design, the perception model function identifier can include one or more of the following: an identifier of a direct perception function or an identifier of an auxiliary perception function. In this way, the perception model function identifier can be explicitly indicated for different perception model functions.

[0012] In a possible design, the direct perception function can include one or more of the following: imaging, positioning, speed measurement, target detection or image segmentation; and / or, the auxiliary perception function can include determining an intermediate variable. In this way, the corresponding perception model registration can be performed for different perception model functions respectively.

[0013] In a possible design, the perception model input information can satisfy at least one of the following: when the perception model is used for imaging, the perception model input information includes one or more of the following: sampling information of an echo signal or a range image;

[0014] When the perception model is used for target detection, the perception model input information includes one or more of the following: sampling information of the echo signal or an imaging result;

[0015] When the perception model is used for speed measurement, the perception model input information includes one or more of the following: channel quality information or a Doppler shift spectrum.

[0016] By the method, the information matched for different perception model functions can be input.

[0017] In a possible design, the perception model performance information can satisfy at least one of the following:

[0018] When the perception model is used for imaging, the perception model performance information includes one or more of the following: an F1-score, an inverse angular distance or a ratio of an intersection to a union;

[0019] When the perception model is used for target detection, the perception model performance information includes one or more of the following: the F1-score, the inverse angular distance or the ratio of the intersection to the union;

[0020] When the perception model is used for speed measurement, the perception model performance information includes one or more of the following: accuracy or mean square error.

[0021] Through the above method, the performance of different perception models can be obtained for different perception model functions.

[0022] In a possible design, the perception model available conditions can satisfy at least one of the following:

[0023] When the perception model is used for imaging, the perception model available conditions include one or more of the following: conditions satisfied by a receptive field and resolution, a threshold of time synchronization error, or a threshold of position synchronization error;

[0024] When the perception model is used for target detection, the perception model available conditions include one or more of the following: conditions satisfied by a receptive field and target size, a threshold of time synchronization error, or a threshold of position synchronization error;

[0025] When the perception model is used for speed measurement, the perception model available conditions include one or more of the following: conditions satisfied by a receptive field and a preset time, or a threshold of time synchronization error.

[0026] Through the above method, different perception models can be applied to different scenarios for different perception model functions.

[0027] In a possible design, first information is received, which can be used to determine the registration information of the perception model. In this way, the registration information of the perception model that needs to be registered can be accurately determined according to the first information.

[0028] In a possible design, the first information can include one or more of the following parameters: a required registration perception model function, the perception model identifier, the perception model input information, the perception model performance information, or the perception model available conditions. In this way, the terminal device can determine whether there is a perception model matching the first information in the local model based on the first information.

[0029] In a possible design, the first information can include a perception model function identifier and a mapping relationship between a perception model function and the perception model function identifier. In this way, the terminal device can select a matching perception model in the local model based on the indication of the network device.

[0030] In a possible design, according to the first information, the perception model matching the first information is determined in the stored model. In this way, the perception model that needs to be registered can be accurately determined based on the first information.

[0031] In a possible design, the first information can include perception model architecture information. In this way, the network device can indicate, for the terminal device, the perception model architecture information of the perception model that needs to be trained, so that the terminal device can accurately train the perception model.

[0032] In a possible design, the perception model architecture information can indicate at least one perception model architecture corresponding to at least one perception model function respectively.

[0033] In a possible design, the perception model requirement is transmitted, and the perception model requirement can be used to determine the perception model architecture. In this way, the network device can determine, for the terminal device, appropriate perception model architecture information based on the perception model requirement of the terminal device.

[0034] In a possible design, the model is trained according to the perception model architecture information, to obtain the perception model. In this way, the terminal device can train an accurate perception model.

[0035] In a possible design, the first information can include a mapping relationship between the perception model function and the perception model architecture information. In this way, the network device can agree on the perception model architecture for the terminal device, so that the terminal device can select appropriate perception model architecture according to its own requirement.

[0036] In a possible design, the perception model architecture information of the perception model is determined according to the mapping relationship and the perception model requirement; and the model is trained according to the perception model architecture information of the perception model, to obtain the perception model. In this way, the terminal device can select appropriate perception model architecture based on its own requirement, and thus train an accurate perception model.

[0037] In a possible design, the perception model requirement can include one or more of the following: a perception model function, perception model input information, and a perception model available condition. In this way, the network device or the terminal device can determine appropriate perception model network architecture based on the perception model requirement.

[0038] In a second aspect, a communication method is provided. The method can be applied to a communication apparatus, which can be a network device or a component (for example, a processor, a chip, a chip system, a circuit, or a functional module) in a network device. The method can include: receiving registration information of a perception model, the registration information of the perception model including one or more of the following: a perception model identifier, perception model performance information, perception model input information, or a perception model available condition.

[0039] By the method, the registration of the perception model can be implemented, so that the network device manages the perception model, and then the perception models of the plurality of terminal devices can be shared in the communication system, and the perception performance of the communication system is improved.

[0040] In a possible design, the perception model identifier can include a perception model function identifier and a perception model architecture identifier, or the perception model identifier includes a perception model function identifier, a perception model architecture identifier and a terminal device identifier. In this way, the perception model identifier can be flexibly represented by a global identifier or a local identifier.

[0041] In a possible design, the perception model function identifier can include one or more of the following: an identifier of a direct perception function or an identifier of an auxiliary perception function. In this way, the perception model function identifier can be explicitly indicated for different perception model functions.

[0042] In a possible design, the direct perception function can include one or more of the following: imaging, positioning tracking, speed measurement, target detection or image segmentation; and / or, the auxiliary perception function includes determining an intermediate variable. In this way, the corresponding perception model registration can be performed for different perception model functions respectively.

[0043] In a possible design, the perception model input information can satisfy at least one of the following:

[0044] When the perception model is used for imaging, the perception model input information includes one or more of the following: sampling information of a return signal or a range image;

[0045] When the perception model is used for target detection, the perception model input information includes one or more of the following: sampling information of the return signal or an imaging result;

[0046] When the perception model is used for speed measurement, the perception model input information includes one or more of the following: channel quality information or a Doppler shift spectrum.

[0047] By the method, the information matched for different perception model functions can be input.

[0048] In a possible design, the perception model performance information can satisfy at least one of the following:

[0049] When the perception model is used for imaging, the perception model performance information includes one or more of the following: an F1-score, an inverse angular distance or a ratio of intersection to union;

[0050] When the perception model is used for object detection, the perception model performance information includes one or more of the following: the F1-score, the IoU, or the ratio of the intersection to the union.

[0051] When the perception model is used for speed measurement, the perception model performance information includes one or more of the following: the accuracy or the mean square error.

[0052] Through the above method, the performance of different perception models can be obtained for different perception model functions.

[0053] In one possible design, the perception model available conditions can satisfy at least one of the following:

[0054] When the perception model is used for imaging, the perception model available conditions include one or more of the following: conditions for satisfying the receptive field and resolution, a threshold for the time synchronization error, or a threshold for the position synchronization error.

[0055] When the perception model is used for object detection, the perception model available conditions include one or more of the following: conditions for satisfying the receptive field and target size, a threshold for the time synchronization error, or a threshold for the position synchronization error.

[0056] When the perception model is used for speed measurement, the perception model available conditions include one or more of the following: conditions for satisfying the receptive field and preset time, or a threshold for the time synchronization error.

[0057] Through the above method, different perception models can be applied to different scenarios for different perception model functions.

[0058] In one possible design, the first information can be used to determine the registration information of the perception model. In this way, the registration information of the perception model that needs to be registered can be accurately determined according to the first information.

[0059] In one possible design, the first information can include one or more of the following parameters: a required perception model function, the perception model identifier, the perception model input information, the perception model performance information, or the perception model available conditions. In this way, the terminal device can determine whether there is a perception model that matches the first information in the local model based on the first information.

[0060] In one possible design, the first information can include a perception model function identifier and a mapping relationship between the perception model function and the perception model function identifier. In this way, the terminal device can select a matching perception model in the local model based on the indication of the network device.

[0061] In a possible design, the first information can include perception model architecture information. In this way, the network device can indicate, for the terminal device, the perception model architecture information of the perception model that needs to be trained, so that the terminal device can accurately train the perception model.

[0062] In a possible design, the perception model architecture information can indicate at least one perception model architecture corresponding to at least one perception model function respectively.

[0063] In a possible design, the perception model requirement can be used to determine the perception model architecture, and the perception model architecture information is determined according to the perception model requirement. In this way, the network device can determine, for the terminal device, appropriate perception model architecture information based on the perception model requirement of the terminal device.

[0064] In a possible design, the perception model requirement can include one or more of the following: a perception model function, the perception model input information, and the perception model available condition. In this way, the network device or the terminal device can determine appropriate perception model network architecture based on the perception model requirement.

[0065] In a possible design, the first information includes a mapping relationship between the perception model function and the perception model architecture information. In this way, the network device can negotiate the perception model architecture for the terminal device, so that the terminal device can select appropriate perception model architecture according to its own requirement.

[0066] In a third aspect, the present application also provides a communication apparatus, which can be a terminal device, or can be a component (for example, a processor, a chip, a chip system, a circuit, or a functional module, etc.) in a terminal device. The communication apparatus has a function of implementing the method in the first aspect or in each possible design example of the first aspect. The function can be implemented by hardware, or can be implemented by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions.

[0067] In a possible design, the structure of the communication apparatus can include a processing unit, and optionally can also include a transceiving unit. These units can perform the functions of the method in the first aspect or in each possible design example of the first aspect, which will not be repeated here.

[0068] In one possible design, the communication apparatus can include one or more processors, and optionally, a memory and / or a transceiver, where the transceiver can be used to receive and / or transmit data, message or information, and to communicate with other devices in the system. The processor(s) can be configured to support the communication apparatus to perform the corresponding functions of the above-described first aspect and / or various possible design examples of the first aspect. The memory can be coupled to the processor(s) and can store program instructions and data for the communication apparatus.

[0069] In one possible design, the communication apparatus can include one or more processors, and optionally, a memory and / or a transceiver, where the transceiver can be used to receive and / or transmit data, message or information, and to communicate with other devices in the system. The processor(s) can be configured to support the communication apparatus to perform the corresponding functions of the above-described first aspect and / or various possible design examples of the first aspect. The memory can be coupled to the processor(s) and can store program instructions and data for the communication apparatus.

[0070] In one possible design, the communication apparatus can include one or more processors, and optionally, a memory and / or a transceiver, where the transceiver can be used to receive and / or transmit data, message or information, and to communicate with other devices in the system. The processor(s) can be configured to support the communication apparatus to perform the corresponding functions of the above-described first aspect and / or various possible design examples of the first aspect. The memory can be coupled to the processor(s) and can store program instructions and data for the communication apparatus.

[0071] In one possible design, the communication apparatus can include one or more processors, and optionally, a memory and / or a transceiver, where the transceiver can be used to receive and / or transmit data, message or information, and to communicate with other devices in the system. The processor(s) can be configured to support the communication apparatus to perform the corresponding functions of the above-described first aspect and / or various possible design examples of the first aspect. The memory can be coupled to the processor(s) and can store program instructions and data for the communication apparatus.

[0072] In one possible design, the communication apparatus can include one or more processors, and optionally, a memory and / or a transceiver, where the transceiver can be used to receive and / or transmit data, message or information, and to communicate with other devices in the system. The processor(s) can be configured to support the communication apparatus to perform the corresponding functions of the above-described first aspect and / or various possible design examples of the first aspect. The memory can be coupled to the processor(s) and can store program instructions and data for the communication apparatus.

[0073] In a sixth aspect, a computer-readable storage medium is provided, which stores program instructions. When the program instructions are run on a computer, the computer is caused to perform the method in the first aspect or any possible implementation of the first aspect, or the method in the second aspect or any possible implementation of the second aspect. Exemplarily, the computer-readable storage medium can be any available medium that can be accessed by a computer. For example, but not limited to: the computer-readable medium can include a non-transitory computer-readable medium, a random-access memory (RAM), a read-only memory (ROM), an electrically EPROM (EEPROM), a CD-ROM or other optical disk storage, a magnetic disk storage medium or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer.

[0074] In a seventh aspect, a computer program product is provided, which includes a computer program or instructions. When the computer program or instructions are run on a computer, the method in the first aspect or any possible implementation of the first aspect, or the method in the second aspect or any possible implementation of the second aspect is performed.

[0075] In an eighth aspect, a chip or chip system is provided, which includes one or more processors coupled with at least one memory for reading and executing program instructions stored in the memory, so that the chip or chip system implements the method in the first aspect or any possible implementation of the first aspect, or the method in the second aspect or any possible implementation of the second aspect.

[0076] The above-mentioned various aspects in the third aspect to the eighth aspect and the technical effects that can be achieved by the various aspects can refer to the technical effects that can be achieved by the first aspect or the various possible implementations of the first aspect, or the technical effects that can be achieved by the second aspect or the various possible implementations of the second aspect, which will not be repeated here. BRIEF DESCRIPTION OF DRAWINGS

[0077] FIG. 1 is a schematic diagram of an architecture of a communication system provided by the present application;

[0078] FIG. 2 is a schematic diagram of a flow of a communication method provided by the present application;

[0079] FIG. 3 is a schematic diagram of a global ID provided by the present application;

[0080] FIG. 4 is a schematic diagram of a flow of an example of a communication method provided by the present application;

[0081] FIG. 5 is a flow diagram of an example of another communication method provided by the present application;

[0082] FIG. 6 is a flow diagram of an example of another communication method provided by the present application;

[0083] FIG. 7 is a flow diagram of an example of another communication method provided by the present application;

[0084] FIG. 8 is a structural diagram of a communication apparatus provided by the present application;

[0085] FIG. 9 is a structural diagram of a communication apparatus provided by the present application. DETAILED DESCRIPTION

[0086] The embodiments of the present application provide a communication method and apparatus for implementing registration of an intelligent sensing model. The method and apparatus provided by the present application are based on the same technical concept, and the implementation of the apparatus and the method can be referred to each other, and the repeated parts will not be described herein.

[0087] In the description of the present application, the terms "first", "second", etc. are used only for the purpose of distinguishing the described objects, and cannot be understood as indicating or implying relative importance, nor indicating or implying sequence.

[0088] In the description of the present application, "at least one" means one or more, and more means two or more. "At least one of the following" or the like means any combination of the items, including any combination of single item or multiple items. For example, at least one of a, b or c can represent a, b, c, a and b, a and c, b and c, or a and b and c, wherein a, b and c can be single or multiple.

[0089] In the description of the present application, "and / or" describes the association relationship between the associated objects, which means that there can be three kinds of relationships, for example, A and / or B can represent the cases of A alone, A and B together, B alone, wherein A and B can be single or multiple. " / " represents "or", for example, a / b represents a or b.

[0090] In order to more clearly describe the technical solutions of the embodiments of the present application, the communication method and apparatus provided by the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0091] The technical solutions in the embodiments of the present application can be applied to various communication systems, such as a universal mobile telecommunications system (UMTS), a wireless local area network (WLAN), a wireless fidelity (Wi-Fi) system, a 4th generation (4G) mobile communication system (such as a long term evolution (LTE) system), a 5th generation (5G) mobile communication system (such as a new radio (NR) system), and a future communication network.

[0092] For example, FIG. 1 shows a schematic diagram of an architecture of a possible communication system to which the embodiments of the present application are applied. As shown in FIG. 1, the communication system 10 can include a radio access network (RAN) 100 and a core network (CN) 200. Optionally, the communication system 10 can also include an Internet 300.

[0093] The RAN 100 includes at least one RAN node (such as 110a and 110b in FIG. 1, collectively referred to as 110) and at least one terminal device (such as 120a-120j in FIG. 1, collectively referred to as 120). The RAN 100 can also include other RAN nodes, such as a wireless relay device and / or a wireless backhaul device (not shown in FIG. 1), etc. The terminal devices 120 are connected to the RAN nodes 110 in a wireless manner. The RAN nodes 110 are connected to the core network 200 in a wireless or wired manner. The core network devices in the core network 200 and the RAN nodes 110 in the RAN 100 can be different physical devices respectively, or can be the same physical device integrated with the logical functions of the core network and the logical functions of the radio access network.

[0094] The RAN 100 can be a 3rd generation partnership project (3GPP) related cellular system, such as a 4G, 5G mobile communication system, or a future-oriented communication system. The RAN 100 can also be an open radio access network (O-RAN or ORAN), a cloud radio access network (CRAN), or a WiFi system. The RAN 100 can also be a communication system in which two or more of the above systems are integrated.

[0095] The RAN node 110, which can also be referred to as a RAN entity or an access node, etc., forms part of the communication system 10 and is configured to facilitate wireless access by terminal devices. The RAN nodes 110 in the communication system 10 can be of the same type or different types. In some scenarios, the roles of a RAN node 110 and a terminal device 120 are relative, e.g., a drone or a helicopter 120i in Figure 1 can be configured to move like a mobile base station, and for a terminal device 120j accessing the RAN 100 via the drone 120i, the drone 120i is a base station; but for a base station 110a, the drone 120i is a terminal device. The RAN nodes 110 and the terminal devices 120 are sometimes referred to as communication apparatuses, e.g., the network elements 110a and 110b in Figure 1 can be understood as communication apparatuses with base station functionality, and the network elements 120a-120j can be understood as communication apparatuses with terminal device functionality.

[0096] The RAN node can also be referred to as a network device. In the following, the network device is used unless specified otherwise.

[0097] In a possible scenario, the network device can also be referred to as an access network device, which can be a base station, an evolved NodeB (eNodeB), an access point (AP), a transmission reception point (TRP), a base station in a future mobile communication system, or an access node in a WiFi system, etc. The access network device can be a macro base station (e.g., 110a in Figure 1), a micro base station or an indoor station (e.g., 110b in Figure 1), a relay node or a donor node, or a wireless controller in a CRAN scenario. Optionally, the access network device can also be a server, a wearable device, a vehicle or a vehicle-mounted device, etc. For example, the access network device in vehicle to everything (V2X) technology can be a road side unit (RSU). All or part of the functions of the access network device in the present application can also be implemented by software functions running on hardware, or by virtualized functions instantiated on a platform (e.g., a cloud platform). The access network device in the present application can also be a logical node, a logical module or software that can implement all or part of the functions of the access network device.

[0098] In another possible scenario, a terminal device accesses a network device through multiple access network devices. The multiple access network devices can be centralized units (CUs), distributed units (DUs), or radio units (RUs). For example, the multiple access network devices can be a CU-CP, a CU-UP, a DU, or a RU. The CU (or CU-CP and CU-UP) and the DU or the RU can be separately configured, or can be included in the same network element, for example, a baseband unit (BBU). The RU can be included in a radio frequency device or a radio frequency unit, for example, a remote radio unit (RRU), an active antenna unit (AAU), or a remote radio head (RRH).

[0099] In different systems, the CU (or CU-CP and CU-UP), the DU, or the RU can also have different names, but those skilled in the art can understand their meanings. For example, in an ORAN system, the CU can also be referred to as an open CU (O-CU), the DU can also be referred to as an open DU (O-DU), the CU-CP can also be referred to as an open CU-CP (O-CU-CP), the CU-UP can also be referred to as an open CU-UP (O-CU-UP), and the RU can also be referred to as an open RU (O-RU). Any of the CU (or CU-CP, CU-UP), the DU, and the RU in this application can be implemented by a software module, a hardware module, or a combination of a software module and a hardware module.

[0100] The terminal device can also be referred to as user equipment (UE), mobile station, mobile terminal, etc. The terminal device can be widely applied to various scenarios, such as device-to-device (D2D), vehicle to everything (V2X) communication, machine-type communication (MTC), internet of things (IOT), virtual reality, augmented reality, industrial control, automatic driving, remote medical treatment, smart grid, smart furniture, smart office, smart wear, smart transportation, smart city, etc. The terminal device can be a mobile phone, tablet computer, computer with wireless transceiver function, wearable device, vehicle, unmanned aerial vehicle, helicopter, airplane, ship, robot, mechanical arm, smart home device, etc. Embodiments of the present application do not limit the device form of the terminal device.

[0101] In some scenarios, the network device can send a downlink signal to the terminal device, and the terminal device can send an uplink signal to the network device. In addition, the network devices can also communicate with each other, and the terminal devices can also communicate with each other.

[0102] The communication system described in the embodiments of the present application is used to more clearly illustrate the technical solutions of the embodiments of the present application, and does not constitute a limitation on the technical solutions provided by the embodiments of the present application. Those skilled in the art can know that, with the evolution of network architecture and the appearance of new business scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0103] At present, how to register the perception model of the terminal device to the network device so that the network device can manage the model after registration is an urgent problem to be solved. Based on this, the embodiments of the present application provide a communication method, which can realize the registration of the perception model.

[0104] In the following embodiments, the communication method provided by the present application is described in detail taking the terminal device and the network device as examples. It should be understood that the operations performed by the terminal device can also be implemented by a processor in the terminal device, or a chip or chip system, or a functional module, etc. The operations performed by the network device can also be implemented by a processor in the network device, or a chip or chip system, or a functional module, etc. The present application does not limit this.

[0105] Based on the above description, the communication method provided by the embodiments of the present application can be referred to FIG. 2. The flow of the method can include:

[0106] Step 201: The terminal device sends the registration information of the perception model, and correspondingly, the network device receives the registration information of the perception model.

[0107] The registration information of the perception model can include one or more of the following: a perception model identifier (model ID), perception model capability information (model capability), perception model input information (model input), or a perception model validity condition.

[0108] In this application, the perception model can also be described as an intelligent perception model, etc.

[0109] In some embodiments, the perception model can include a terminal device-side model (UE-side model), a network-side model (network-side model), and a two-sided model (two-sided model). The terminal device-side model is an AI or ML model whose training or inference is performed on a terminal device. The network-side model is an AI or ML model whose training or inference is performed on a network device. The two-sided model is an AI or ML model whose training or inference is jointly performed by a terminal device and a network device, that is, part of the training or inference is performed by the terminal device and the remaining part is performed by the network device, or part of the training or inference is performed by the network device and the remaining part is performed by the terminal device.

[0110] In some embodiments, the perception model registered by the terminal device to the network device can be a terminal device-side model, or a model part of a two-sided model obtained by the terminal device performing training or inference, which is not limited in this application.

[0111] In an optional implementation, the perception model identifier can include a perception model function identifier and a perception model architecture identifier. Alternatively, the perception model identifier can also include a perception model function identifier, a perception model architecture identifier, and a terminal device identifier.

[0112] When the perception model identifier includes a perception model function identifier and a perception model architecture identifier, it can be understood that the perception model identifier is a local identifier (local ID). When the perception model identifier includes a perception model function identifier, a perception model architecture identifier, and a terminal device identifier, it can be understood that the perception model identifier is a global identifier (global ID).

[0113] In some scenarios, the terminal device can be accurately identified without indicating the terminal device identifier, and the perception model identifier can use a local ID in the registration process of the perception model.

[0114] In other scenarios, if the terminal device identifier is not indicated, the terminal device cannot be accurately identified, and the perception model identifier can use a global ID in the registration process of the perception model.

[0115] For example, the format of the global ID can be shown in FIG. 3.

[0116] The terminal device identifier can be used to uniquely identify the terminal device. For example, the terminal device identifier can include one or more of the following: an international mobile subscriber identity (IMSI), a 5G globally unique temporary identity (5G-GUTI), and the like.

[0117] In some embodiments, the perception model function identifier can include one or more of the following: an identifier of a direct perception function or an identifier of an auxiliary perception function.

[0118] The direct perception function means that the perception result can be directly output based on the perception model.

[0119] For example, the direct perception function can include one or more of the following: imaging, positioning, speed measurement, target detection, image segmentation, and the like.

[0120] The auxiliary perception function means that an intermediate variable is output based on the perception model, which can also be understood as the auxiliary perception function including determining an intermediate variable.

[0121] For example, the perception model based on the auxiliary perception function can output a new measurement result or an enhanced result of an existing measurement result, such as an intermediate variable including clutter recognition, suppression, scatter point statistical information, Doppler shift, channel state information, and the like.

[0122] In an optional implementation, the perception model architecture and the perception model function are associated with each other. For example, for a perception model function, one or more perception model architectures can be used based on different available conditions of the perception model. For example, for the imaging function of the perception model, a convolutional neural network (CNN) model architecture with different receptive fields and channel numbers can be used. Each perception model architecture can be represented by perception model architecture description information and / or a perception model architecture identifier, wherein the perception model architecture description information can include one or more of the following: perception model architecture content, an identifier of the perception model architecture content. For example, the correspondence between a perception model function and a perception model architecture can be shown in Table 1 as follows.

[0123] Table 1

[0124] It should be understood that the perception model architecture description information and the perception model architecture identifier in Table 1 can both be included or one of them can be included, and the present application does not limit this.

[0125] In an example, for the imaging function of the direct perception function, the corresponding perception model architecture is perception model architecture N1, perception model architecture N2, …, and perception model architecture Nk. Taking the CNN model architecture as the perception model architecture N1 as an example, the perception model architecture N1 description information can include one or more of the following:

[0126] 1) Input layer: receives image data, which can be a color image with a size of 32x32x3 (i.e., 32 pixels in width, 32 pixels in height, and 3 channels).

[0127] 2) Convolutional layer 1: the convolution kernel size is 3x3, and the stride is 1; the number of channels of the output feature map is 64; a rectified linear unit (ReLU) activation function is used for nonlinear transformation.

[0128] 3) Pooling layer 1: maximum pooling operation, with a pooling window size of 2x2 and a stride of 2.

[0129] 4) Convolutional layer 2: the convolution kernel size is 3x3, and the stride is 1; the number of channels of the output feature map is 128; a ReLU activation function is used for nonlinear transformation.

[0130] 5) Pooling layer 2: maximum pooling operation, with a pooling window size of 2x2 and a stride of 2.

[0131] 6) Flatten layer: flattens the output feature map of the last pooling layer into a one-dimensional vector.

[0132] 7) Fully connected layer 1: the number of output nodes is 2; a softmax activation function is used for binary classification, representing the imaging result.

[0133] It should be understood that the above description information is only an example, and the description information of a perception model architecture can also be represented in other ways, and the present application does not limit this.

[0134] Optionally, the terminal device determines the perception model architecture corresponding to each perception model function by itself. For example, the correspondence indicated in the foregoing Table 1 can be determined by the network device.

[0135] Optionally, in order to reduce the complexity of model management of the network device, the network device can determine the perception model architecture corresponding to each perception model function. For example, the correspondence indicated in the foregoing Table 1 can be determined by the network device.

[0136] In a possible manner, the terminal device can encode the perception model function identifier and the perception model architecture identifier of the perception model by using A and B bits respectively according to the number of perception model functions and the number of perception model architectures that the terminal device already has. A is the number of bits used to indicate the number of perception model functions, B is the number of bits used to indicate the number of perception model architectures, and A and B are positive integers. For example, if the number of perception model functions that the terminal device already has is 4 and the number of perception model architectures is 2, the number of perception model functions can be indicated by A=2 bits and the number of perception model architectures can be indicated by B=1 bit. Therefore, the terminal device can encode the perception model function identifier and the perception model architecture identifier by using A+B=3 bits to indicate the perception model function and the perception model architecture of the perception model. For example, “100” can represent the positioning function and the CNN architecture 1, where “10” can be understood as the identifier of the positioning function, and “0” can be understood as the identifier of the CNN architecture 1. For example, the CNN architecture 1 can be the perception model architecture N1 described above, and reference can be made to the description information of the foregoing perception model architecture N1, which is not described herein again.

[0137] In a possible manner, the network device can determine the model function identifier and indicate the mapping relationship between the perception model function and the model function identifier.

[0138] For example, the mapping relationship between the perception model function and the perception model function identifier can be as shown in Table 2.

[0139] Table 2

[0140] In some embodiments, the perception model input information can include one or more of the following: samples of echo signals, range images, imaging results, or the number of input channels, and the like.

[0141] The range image is a one-dimensional distribution image of a target scattering point in a specific view angle, and can be a result obtained by performing pulse compression on the echo signal.

[0142] The perception model performance information can include the perception accuracy of the model and the like.

[0143] The available condition of the perception model can represent how many features the perception model can extract in a spatial and temporal neighborhood. Optionally, the available condition of the perception model can include one or more of the following: a model condition or a scene condition.

[0144] For example, the model conditions can include one or more of the following: receptive field, number of channels or heads, etc. The scenario conditions can include one or more of the following: time synchronization error, position synchronization error, etc.

[0145] In an alternative embodiment, the registration information can be implemented by a data structure as shown in Table 3, and the registration information can include one or more of the items in Table 3. Table 3 can also be understood as a mapping structure for describing the mapping between the machine learning parameters (the content shown in the left column of Table 3) and the perception parameters (the content shown in the right column of Table 3) of the perception model.

[0146] Table 3

[0147] In some embodiments, the perception model functions differently, and the corresponding perception parameters of the perception model are also different, i.e., the registration information is different.

[0148] For example, when the perception model is used for imaging, the perception model input information can include one or more of the following: sampling information or range image of echo signals; the perception model performance information can include one or more of the following: F1-score or chamfer distance (CD) or Intersection over Union (IoU); the available conditions of the perception model can include one or more of the following: conditions for satisfying the receptive field and resolution, threshold of time synchronization error or threshold of position synchronization error.

[0149] In an example, when the function of the perception model is imaging, the data structure of the registration information can be as shown in Table 4.

[0150] Table 4

[0151] For another example, when the perception model is used for target detection, the perception model input information can include one or more of the following: sampling information or imaging results of echo signals; the perception model performance information can include one or more of the following: F1-score, chamfer distance or Intersection over Union; the available conditions of the perception model can include one or more of the following: conditions for satisfying the receptive field and target size (the target size can also be understood as the size of the target object), threshold of time synchronization error or threshold of position synchronization error.

[0152] In an example, when the function of the perception model is target detection, the data structure of the registration information can be as shown in Table 5.

[0153] Table 5

[0154] For example, when the perception model is used for speed measurement, the perception model input information can include one or more of the following: channel quality information or a doppler frequency shift (DFS) spectrum; the perception model performance information includes one or more of the following: accuracy or mean square error; the perception model available conditions include one or more of the following: conditions for which the receptive field and the preset time are satisfied or a threshold for the time synchronization error.

[0155] The DFS spectrum refers to time-frequency analysis of received channel state information to obtain a graph of the observed signal phase change over time.

[0156] In an example, when the perception model function is speed measurement, the data structure of the registration information can be as shown in Table 6.

[0157] Table 6

[0158] It should be understood that Tables 3-6 are only examples, and the alternative table can also include more or less information, which is not limited in the present application.

[0159] In some embodiments, before the terminal device sends the registration information of the perception model, the terminal device can also perform step 200: the terminal device receives first information, the first information being used to determine the registration information of the perception model. Correspondingly, the network device sends the first information.

[0160] Optionally, the first information can indicate the perception model parameters, which can be used to determine the registration information of the perception model. It can also be understood that the first information can include the perception model parameters.

[0161] In some embodiments a1, the first information can include one or more of the following parameters: a required registration perception model function, the perception model identifier, the perception model input information, the perception model performance information, or the perception model available conditions.

[0162] The required registration perception model function can be understood as the perception model function of the perception model supported by the network device for the terminal device to register, for example, the perception model function of the perception model supported by the network device for the terminal device to register is imaging.

[0163] In this embodiment a1, before the terminal device registers the perception model, the terminal device can train and store at least one perception model. It can also be understood that the perception model architecture of the perception model trained by the terminal device is determined by the terminal device itself. The terminal device can also encode the perception model function identifier and the perception model architecture identifier of the perception model using A and B bits, respectively, according to the number of perception model functions and the number of perception model architectures it already has.

[0164] In this embodiment a1, the first information can also be understood as perception model registration indication information. When the terminal device receives the first information, according to the first information, in the stored model, the perception model matching the first information is determined. That is, the terminal device can determine whether there is a perception model that needs to be registered in the stored model according to the first information. When the terminal device determines that there is a perception model that needs to be registered, the registration information is sent to the network device.

[0165] The process of determining the perception model matching the first information by the terminal device can be understood as that the terminal device determines that there is a perception model matching each item of the perception model parameters indicated by the first information.

[0166] Optionally, after the network device receives the registration information sent by the terminal device, the confirmation information can be sent to the terminal device to notify the terminal device that the perception model registration is successful.

[0167] In some embodiments a2, the first information can include the perception model function identifier and the mapping relationship between the perception model function and the perception model function identifier (for example, as shown in the foregoing Table 2).

[0168] Optionally, the first information can also include one or more of the following parameters: the required registration perception model function, the perception model identifier, the perception model input information, the perception model performance information, or the perception model available condition.

[0169] In this embodiment a2, the terminal device can train at least one perception model and store it before performing the perception model registration. It can also be understood that the perception model architecture of the perception model trained by the terminal device is determined by the terminal device itself. The network device can indicate the perception model function identifier through the first information.

[0170] In this embodiment a2, the first information can also be understood as perception model registration indication information. When the terminal device receives the first information, according to the first information, in the stored model, the perception model matching the first information is determined.

[0171] For example, the terminal device can determine the perception model function according to the perception model function identifier and the mapping relationship between the perception model function and the perception model function identifier indicated by the first information, and then the terminal device determines the perception model that needs to be registered according to the perception model function and the perception model architecture determined by the terminal device itself, and then the terminal device sends the registration information to the network device.

[0172] Optionally, after the network device receives the registration information sent by the terminal device, the confirmation information can be sent to the terminal device to notify the terminal device that the perception model registration is successful.

[0173] In some embodiments a3, the first information can comprise perception model architecture information.

[0174] In an example, the perception model architecture information can indicate at least one perception model architecture corresponding to at least one perception model function respectively.

[0175] Optionally, the perception model architecture information can comprise perception model architecture description information and / or perception model architecture identifier.

[0176] In an optional implementation, the terminal device can send a perception model requirement before receiving the first information, the perception model requirement being used to determine the perception model architecture, accordingly, the network device receives the perception model requirement and determines the perception model architecture information according to the perception model requirement.

[0177] It can also be understood that the terminal device sends the perception model requirement to the network device to request the perception model architecture.

[0178] For example, the perception model requirement can comprise one or more of the following: perception model function, perception model input information, and perception model available condition.

[0179] For example, taking imaging as an example of the perception model function, the perception model requirement can comprise one or more of the contents shown in Table 7 as follows:

[0180] Table 7

[0181] For example, the receptive field requirement can be that the receptive field is greater than or equal to the resolution. The time synchronization error requirement can be that the time synchronization error is less than a threshold value of the time synchronization error. The position synchronization error requirement can be that the position synchronization error is less than a threshold value of the position synchronization error.

[0182] Optionally, when the network device determines the perception model architecture information according to the perception model requirement, the network device can determine the matching perception model architecture information according to the perception model requirement. For example, when the perception model requirement indicates the imaging function, the network device can determine the perception model architecture information corresponding to the imaging function.

[0183] In this embodiment a3, in a possible manner, after receiving the first information, the terminal device can send response information of the first information to the network device to indicate that the perception model architecture information is successfully received.

[0184] In an optional embodiment, the terminal device can train a model according to the perception model architecture information to obtain a perception model. It can be understood that the process of training a model by the terminal device according to the perception model architecture information can also be understood as the process of locally training and updating model parameters by the terminal device.

[0185] In this embodiment a3, it can be understood that the network device determines the perception model network architecture of the perception model for the terminal device, so that the terminal device trains a model based on this.

[0186] After the terminal device trains a perception model, the terminal device can send registration information of the perception model to the network device.

[0187] In an example, since the terminal device has sent the perception model requirement information, the terminal device can include at least one of Table 8 when sending the registration information.

[0188] Table 8

[0189] Similarly, in this embodiment a3, after the network device receives the registration information sent by the terminal device, the network device can send confirmation information to the terminal device to notify the terminal device that the perception model registration is successful.

[0190] In some embodiments a4, the first information can include a mapping relationship between the perception model function and the perception model architecture information (for example, which can be as shown in Table 1 described above).

[0191] In this embodiment a4, the network device can provide the terminal device with multiple perception model architecture information, and the terminal device can determine a suitable perception model architecture based on the perception model architecture information provided by the network device based on its own perception model requirement.

[0192] For example, the terminal device can determine the perception model architecture information of the perception model according to the mapping relationship and the perception model requirement, and train a model according to the perception model architecture information of the perception model to obtain the perception model.

[0193] The perception model requirement can refer to the related description described above, which will not be described here.

[0194] For example, when the perception model requirement indicates an imaging function, the terminal device can select the perception model architecture information corresponding to the imaging function from the multiple perception model architecture information provided by the network device when determining the perception model architecture information of the perception model according to the mapping relationship and the perception model requirement.

[0195] After the terminal device trains a perception model, the terminal device can send registration information of the perception model to the network device.

[0196] Similarly, in this embodiment a4, after the network device receives the registration information sent by the terminal device, the network device can send confirmation information to the terminal device to notify the terminal device that the perception model registration is successful.

[0197] Through the above communication method, the registration of the perception model can be realized, so that the network device manages the perception model, and then the perception models of multiple terminal devices can be shared in the communication system, and the perception performance of the communication system is improved.

[0198] Based on the above embodiments, the communication method provided by the embodiments of the present application is described in detail below through the examples shown in FIGS. 4-7.

[0199] FIG. 4 shows an example of a communication method. In this example, the terminal device can train and store at least one perception model before performing perception model registration. It can also be understood that the perception model architecture of the perception model trained by the terminal device is determined by the terminal device itself. The terminal device can also encode the perception model function identifier and the perception model architecture identifier of the perception model according to the number of existing perception model functions and the number of perception model architectures, respectively, using A and B bits. For example, the flow of this example can include the following steps.

[0200] Step 401: The network device sends first information to the terminal device.

[0201] The first information can be understood as perception model registration indication information.

[0202] The first information can include one or more of the following parameters: required registration perception model function, perception model identifier, perception model input information, perception model performance information, or perception model available condition.

[0203] The specific content of each item of information can be referred to the related description in the foregoing embodiments, which will not be described here.

[0204] Step 402: The terminal device determines whether there is a perception model that needs to be registered according to the first information, and if so, performs step 403.

[0205] It can also be understood that the terminal device determines the perception model matching the first information in the stored model according to the first information.

[0206] Step 403: The terminal device sends registration information of the perception model to the network device.

[0207] The registration information can be referred to the related description in the foregoing embodiments, which will not be described here.

[0208] Step 404: The network device sends confirmation information to the terminal device.

[0209] The confirmation information is used to notify the terminal device that the perception model is successfully registered.

[0210] Alternatively, the confirmation information can also be described as a model registration success confirmation, etc.

[0211] Based on the above examples, the registration of the perception model can be implemented, so that the network device manages the perception model, and thus the perception models of multiple terminal devices can be shared in the communication system, and the perception performance of the communication system is improved.

[0212] FIG. 5 shows another example of a communication method. In this example, the terminal device can train at least one perception model and store it before registering the perception model. It can also be understood that the perception model architecture of the perception model trained by the terminal device is determined by the terminal device itself. The network device can indicate the perception model function identifier to the terminal device. Exemplarily, the flow of this example can include the following steps.

[0213] Step 501: The network device sends first information to the terminal device.

[0214] The first information can be understood as perception model registration indication information.

[0215] The first information can include a perception model function identifier and a mapping relationship between the perception model function and the perception model function identifier (for example, as shown in Table 2 described above).

[0216] Step 502: The terminal device determines a perception model matching the first information from the stored model according to the first information.

[0217] Exemplarily, the terminal device can determine the perception model function according to the perception model function identifier indicated by the first information and the mapping relationship between the perception model function and the perception model function identifier, and then the terminal device determines the perception model to be registered according to the perception model function and the perception model architecture determined by the terminal device itself.

[0218] Alternatively, the terminal device can determine the perception model identifier of the perception model according to the mapping relationship and the perception model architecture determined by the terminal device itself. The perception model identifier can be a global identifier or a local identifier.

[0219] Step 503: The terminal device sends registration information of the perception model to the network device.

[0220] The registration information can refer to the related description in the foregoing embodiments, which will not be described here.

[0221] Step 504: The network device sends confirmation information to the terminal device.

[0222] The confirmation information is in response to the registration information, and is used to notify the terminal device that the perception model is successfully registered.

[0223] Alternatively, the confirmation information can also be described as model registration success confirmation, etc.

[0224] Based on the above examples, the registration of the perception model can be implemented, so that the network device manages the perception model, and thus the perception models of multiple terminal devices can be shared in the communication system, and the perception performance of the communication system is improved.

[0225] FIG. 6 shows another example of a communication method. In this example, the network device determines the perception model network architecture of the perception model for the terminal device, so that the terminal device trains the model based on it. It can also be understood that the network device agrees on the perception model architecture and indicates the perception model architecture that needs to be trained for the terminal device. Illustratively, the flow of this example can include the following steps.

[0226] Step 601: The terminal device sends a perception model requirement to the network device.

[0227] The perception model requirement can refer to the related description related to the foregoing embodiments, which will not be repeated here.

[0228] Step 602: The network device determines the perception model architecture information according to the perception model requirement.

[0229] The perception model architecture information can refer to the related description related to the foregoing embodiments, which will not be repeated here.

[0230] Step 603: The network device sends first information to the terminal device, and the first information can include the perception model architecture information.

[0231] Step 604: The terminal device sends response information of the first information to the network device.

[0232] The response information of the first information can indicate that the perception model architecture information is successfully received.

[0233] Alternatively, the response information of the first information can also be described as model architecture confirmation information.

[0234] Step 605: The terminal device trains the model according to the perception model architecture information included in the first information to obtain the perception model.

[0235] It can be understood that the process of training the model according to the perception model architecture information by the terminal device can also be understood as the process of updating the model parameters by the terminal device for local training.

[0236] Step 606: The terminal device sends registration information of the perception model to the network device.

[0237] The registration information can be referred to the related description in the foregoing embodiments, and details are not described herein.

[0238] For example, the registration information can be referred to Table 8.

[0239] Step 607: The network device sends confirmation information to the terminal device.

[0240] The confirmation information is in response to the registration information, and the confirmation information is used to notify the terminal device that the registration of the perception model is successful.

[0241] Alternatively, the confirmation information can also be described as model registration success confirmation, etc.

[0242] Based on the foregoing example, the registration of the perception model can be implemented, so that the network device manages the perception model, and then the perception models of multiple terminal devices can be shared in the communication system, and the perception performance of the communication system is improved.

[0243] FIG. 7 shows another example of a communication method. In this example, the network device can provide the terminal device with a plurality of perception model architecture information, and the terminal device can determine a suitable perception model architecture based on its own perception model requirements and the perception model architecture information provided by the network device. It can also be understood that the network device agrees on a perception model architecture, and indicates the agreed perception model architecture to the terminal device, and the terminal device can select a matching perception model architecture in the agreed perception model architecture according to the perception model requirements (which can also be understood as service requirements). For example, the flow of this example can include the following steps.

[0244] Step 701: The network device sends first information to the terminal device, and the first information can include a mapping relationship between the perception model function and the perception model architecture information (for example, which can be shown in Table 1).

[0245] Step 702: The terminal device determines the perception model architecture information of the perception model according to the mapping relationship and the perception model requirements.

[0246] The perception model requirements can be referred to the related description in the foregoing embodiments, and details are not described herein.

[0247] Step 703: The terminal device trains the model according to the perception model architecture information of the perception model, and obtains the perception model.

[0248] It can be understood that the process of training the model according to the perception model architecture information by the terminal device can also be understood as the process of updating the model parameters by the terminal device for local training.

[0249] Step 704: The terminal device sends registration information of the perception model to the network device.

[0250] The registration information can refer to the related description in the foregoing embodiments, and details are not described herein.

[0251] Step 705: The network device sends confirmation information to the terminal device.

[0252] The confirmation information is in response to the registration information, and the confirmation information is used to notify the terminal device that the registration of the perception model is successful.

[0253] Alternatively, the confirmation information can also be described as model registration success confirmation, etc.

[0254] Based on the foregoing examples, the registration of the perception model can be implemented, so that the network device manages the perception model, and then the perception models of multiple terminal devices can be shared in the communication system, and the perception performance of the communication system is improved.

[0255] Based on the foregoing embodiments, the embodiments of the present application further provide a communication apparatus. Referring to FIG. 8, the communication apparatus 800 can include a transceiver unit 801 and a processing unit 802. The transceiver unit 801 is configured to perform communication of the communication apparatus 800, such as receiving information (signals or data) or sending information (signals or data). The processing unit 802 is configured to control and manage the actions of the communication apparatus 800. The processing unit 802 can also control the steps performed by the transceiver unit 801.

[0256] For example, the communication apparatus 800 can be the terminal device, the processor of the terminal device, a chip, a chip system, a functional module, or the like in the foregoing embodiments. Alternatively, the communication apparatus 800 can be the network device, the processor of the network device, a chip, a chip system, a functional module, or the like in the foregoing embodiments.

[0257] In one embodiment, when the communication apparatus 800 is used to implement the functions of the terminal device in the foregoing embodiments, the transceiver unit 801 can be configured to send registration information of the perception model, and the registration information of the perception model includes one or more of the following: a perception model identifier, perception model performance information, perception model input information, or a perception model available condition. The processing unit 802 can be configured to control the operation of the transceiver unit 801.

[0258] For example, the perception model identifier can include a perception model function identifier and a perception model architecture identifier, or the perception model identifier includes the perception model function identifier, the perception model architecture identifier, and a terminal device identifier.

[0259] Optionally, the perception model function identifier can comprise one or more of: an identifier of a direct perception function or an identifier of an auxiliary perception function.

[0260] For example, the direct perception function can comprise one or more of: imaging, localization, velocity estimation, object detection or image segmentation; and / or, the auxiliary perception function comprises determining an intermediate variable.

[0261] In an example, the perception model input information can satisfy at least one of:

[0262] When the perception model is used for imaging, the perception model input information comprises one or more of: sampling information of echo signals or range images;

[0263] When the perception model is used for object detection, the perception model input information comprises one or more of: sampling information of the echo signals or imaging results;

[0264] When the perception model is used for velocity estimation, the perception model input information comprises one or more of: channel quality information or Doppler shift spectrograms.

[0265] In an example, the perception model performance information can satisfy at least one of:

[0266] When the perception model is used for imaging, the perception model performance information comprises one or more of: an F1-score, an inverse angular distance or a ratio of intersection over union;

[0267] When the perception model is used for object detection, the perception model performance information comprises one or more of: the F1-score, the inverse angular distance or the ratio of intersection over union;

[0268] When the perception model is used for velocity estimation, the perception model performance information comprises one or more of: accuracy or mean squared error.

[0269] In an example, the perception model usability condition can satisfy at least one of:

[0270] When the perception model is used for imaging, the perception model usability condition comprises one or more of: a condition that receptive fields and resolution are satisfied, a threshold of time synchronization error or a threshold of position synchronization error;

[0271] When the perception model is used for object detection, the perception model usability condition comprises one or more of: a condition that receptive fields and object size are satisfied, the threshold of time synchronization error or the threshold of position synchronization error;

[0272] When the perception model is used for speed measurement, the available conditions of the perception model can include one or more of the following: a condition that a receptive field and a preset time are satisfied or a threshold of the time synchronization error.

[0273] In some embodiments, the transceiver 801 can also be configured to receive first information, which is used to determine registration information of the perception model.

[0274] Optionally, the first information can include one or more of the following parameters: a required registration perception model function, the perception model identifier, the perception model input information, the perception model performance information, or the available conditions of the perception model.

[0275] In an example, the first information can include a perception model function identifier and a mapping relationship between the perception model function and the perception model function identifier.

[0276] The processing unit 802 can also be configured to determine, from the stored models, the perception model that matches the first information according to the first information.

[0277] In another example, the first information can include perception model architecture information.

[0278] For example, the perception model architecture information can indicate at least one perception model architecture corresponding to at least one perception model function, respectively.

[0279] In an optional implementation, the transceiver 801 can also be configured to send a perception model requirement, which is used to determine a perception model architecture.

[0280] Optionally, the processing unit 802 can also be configured to train a model according to the perception model architecture information to obtain the perception model.

[0281] In yet another example, the first information can include a mapping relationship between the perception model function and the perception model architecture information.

[0282] The processing unit 802 can also be configured to determine, from the mapping relationship and a perception model requirement, the perception model architecture information of the perception model; and train a model according to the perception model architecture information of the perception model to obtain the perception model.

[0283] In some embodiments, the perception model requirement can include one or more of the following: a perception model function, the perception model input information, or the available conditions of the perception model.

[0284] In another embodiment, the communication apparatus 800 is configured to implement the functions of the network device in the above embodiments, the transceiver unit 801 can be configured to receive registration information of a perception model, the registration information of the perception model comprising one or more of: a perception model identifier, perception model performance information, perception model input information, or perception model available conditions. The processing unit 802 can be configured to control the operation of the transceiver unit 801.

[0285] For example, the perception model identifier can comprise a perception model function identifier and a perception model architecture identifier, or the perception model identifier comprises a perception model function identifier, a perception model architecture identifier, and a terminal device identifier.

[0286] Optionally, the perception model function identifier can comprise one or more of: an identifier of a direct perception function or an identifier of an auxiliary perception function.

[0287] For example, the direct perception function can comprise one or more of: imaging, localization tracking, velocity estimation, object detection, or image segmentation; and / or, the auxiliary perception function can comprise determining intermediate variables.

[0288] In an example, the perception model input information can satisfy at least one of:

[0289] When the perception model is used for imaging, the perception model input information comprises one or more of: sampling information of echo signals or range images;

[0290] When the perception model is used for object detection, the perception model input information comprises one or more of: sampling information of echo signals or imaging results;

[0291] When the perception model is used for velocity estimation, the perception model input information comprises one or more of: channel quality information or Doppler shift spectrograms.

[0292] In an example, the perception model performance information can satisfy at least one of:

[0293] When the perception model is used for imaging, the perception model performance information comprises one or more of: an F1-score, an inverse angular distance, or a ratio of intersection over union;

[0294] When the perception model is used for object detection, the perception model performance information comprises one or more of: the F1-score, the inverse angular distance, or the ratio of intersection over union;

[0295] When the perception model is used for velocity estimation, the perception model performance information comprises one or more of: accuracy or mean squared error.

[0296] In an example, the perception model availability condition can satisfy at least one of the following:

[0297] When the perception model is used for imaging, the perception model availability condition comprises one or more of the following: a condition that a receptive field and a resolution are satisfied, a threshold of a time synchronization error or a threshold of a position synchronization error;

[0298] When the perception model is used for object detection, the perception model availability condition comprises one or more of the following: a condition that a receptive field and an object size are satisfied, the threshold of the time synchronization error or the threshold of the position synchronization error;

[0299] When the perception model is used for speed measurement, the perception model availability condition comprises one or more of the following: a condition that a receptive field and a preset time are satisfied or the threshold of the time synchronization error.

[0300] In an alternative implementation, the transceiver 801 can also be configured to send first information, the first information being used to determine registration information of the perception model.

[0301] In some embodiments, the first information can comprise one or more of the following parameters: a required registration perception model function, the perception model identifier, the perception model input information, the perception model performance information or the perception model availability condition.

[0302] In other embodiments, the first information can comprise a perception model function identifier and a mapping relationship between the perception model function and the perception model function identifier.

[0303] In yet other embodiments, the first information can comprise perception model architecture information.

[0304] For example, the perception model architecture information can indicate at least one perception model architecture corresponding to at least one perception model function respectively.

[0305] In a possible way, the transceiver 801 can also be configured to receive a perception model requirement, the perception model requirement being used to determine a perception model architecture; and the processing unit 802 can also be configured to determine the perception model architecture information according to the perception model requirement.

[0306] Optionally, the perception model requirement can comprise one or more of the following: a perception model function, the perception model input information, the perception model availability condition.

[0307] In yet other embodiments, the first information can comprise a mapping relationship between the perception model function and the perception model architecture information.

[0308] It should be noted that the division of the units in the embodiments of the present application is illustrative, and is only a logical function division. In actual implementation, another division manner can be used. The function units in the embodiments of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software function unit.

[0309] When the integrated unit is realized in the form of a software function unit and sold or used as an independent product, the integrated unit can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application can be embodied in the form of a software product in essence or in the form of a software product that contributes to the prior art or the whole or part of the technical solutions. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to execute all or part of the steps of the method described in the embodiments of the present application. The foregoing storage medium includes a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0310] Based on the above embodiments, the embodiments of the present application further provide a communication device. Referring to FIG. 9, the communication device 900 can include one or more processors 902. Optionally, the communication device 900 can further include one or more transceivers 901. Optionally, the communication device 900 can further include at least one memory 903. The memory 903 can be arranged inside the communication device 900, or arranged outside the communication device 900. The processor 902 can control the transceiver 901 to receive and send information, messages, or data.

[0311] Specifically, the processor 902 can be a central processing unit (CPU), a network processor (NP), or a combination thereof. The processor 902 can further include a hardware chip. The hardware chip can be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The PLD can be a complex programmable logic device (CPLD), a field-programmable gate array (FPGA), a generic array logic (GAL), or any combination thereof.

[0312] The transceiver 901, the processor 902, and the memory 903 are connected to each other. Optionally, the transceiver 901, the processor 902, and the memory 903 are connected to each other through a bus 904. The bus 904 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in FIG. 9, but it does not mean that there is only one bus or only one type of bus.

[0313] In an optional implementation, the memory 903 is configured to store programs, etc. Specifically, the programs can include program codes including computer operation instructions. The memory 903 can include a RAM, and can also include a non-volatile memory such as one or more disk memories. The processor 902 executes the application programs stored in the memory 903 to implement the above functions, thereby implementing the functions of the communication apparatus 900.

[0314] For example, the communication apparatus 900 can specifically implement the functions of the terminal device or the network device in the above embodiments.

[0315] In an embodiment, when the communication apparatus 900 implements the functions of the terminal device in the foregoing method embodiments, the transceiver 901 can implement the transceiving operations performed by the terminal device in the foregoing method embodiments; the processor 902 can implement operations other than the transceiving operations performed by the terminal device in the foregoing method embodiments. For specific details, refer to the related description in the foregoing method embodiments, which will not be described in detail here.

[0316] In another embodiment, when the communication apparatus 900 implements the functions of the terminal device in the foregoing method embodiments, the processor 902 can implement the operations performed by the terminal device in the foregoing method embodiments. For specific details, refer to the related description in the foregoing method embodiments, which will not be described in detail here.

[0317] In yet another embodiment, when the communication apparatus 900 implements the functions of the network device in the foregoing method embodiments, the transceiver 901 can implement the transceiving operations performed by the network device in the foregoing method embodiments; the processor 902 can implement operations other than the transceiving operations performed by the network device in the foregoing method embodiments. For specific details, refer to the related description in the foregoing method embodiments, which will not be described in detail here.

[0318] In yet another embodiment, when the communication apparatus 900 implements the functions of the network device in the foregoing method embodiments, the processor 902 can implement the operations performed by the network device in the foregoing method embodiments. For specific details, refer to the related description in the foregoing method embodiments, which will not be described in detail here.

[0319] Based on the foregoing embodiments, the embodiments of the present application provide a communication system, which can include the terminal device and the network device and the like involved in the foregoing embodiments.

[0320] The embodiments of the present application further provide a computer readable storage medium for storing a computer program or instructions, which, when executed by a computer, can implement the communication method provided by the foregoing method embodiments.

[0321] The embodiments of the present application further provide a computer program product for storing a computer program or instructions, which, when executed by a computer, can implement the communication method provided by the foregoing method embodiments.

[0322] The embodiments of the present application further provide a chip or chip system, which includes a logic circuit for executing the communication method provided by the foregoing method embodiments.

[0323] The embodiments of the present application also provide a chip or chip system, comprising one or more processors coupled with at least one memory, for invoking a program in the memory to cause the chip or chip system to implement the communication method provided by the above method embodiments.

[0324] The embodiments of the present application also provide a chip or chip system coupled with at least one memory, for implementing the communication method provided by the above method embodiments.

[0325] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) containing computer usable program code.

[0326] The present application is described with reference to the flowcharts and / or block diagrams of the methods, apparatus (systems) and computer program products according to this application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as a combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing apparatus produce the functions specified in one or more flows in the flowcharts and / or one or more blocks in the block diagrams.

[0327] These computer program instructions can also be stored in a computer readable memory that can direct the computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including instruction means, which implements the functions specified in one or more flows in the flowcharts and / or one or more blocks in the block diagrams.

[0328] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide steps for implementing the functions specified in one or more flows in the flowcharts and / or one or more blocks in the block diagrams.

[0329] Obviously, many modifications and variations of the present application are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the appended claims and their equivalents, the application can be practiced otherwise than as specifically described.

Claims

1. A communication method characterized by comprising: The method comprises: sending registration information of the perception model, the registration information of the perception model comprising one or more of: a perception model identifier, perception model performance information, perception model input information, or a perception model available condition.

2. The method of claim 1, wherein, The method further comprises: receiving first information, the first information being used to determine the registration information of the perception model.

3. A communication method characterized by comprising: The method comprises: receiving registration information of the perception model, the registration information of the perception model comprising one or more of: a perception model identifier, perception model performance information, perception model input information, or a perception model available condition.

4. The method of claim 3, wherein, The method further comprises: sending first information, the first information being used to determine the registration information of the perception model.

5. The method according to any one of claims 1 to 4, characterized in that, The perception model identifier comprises a perception model function identifier and a perception model architecture identifier, or the perception model identifier comprises the perception model function identifier, the perception model architecture identifier, and a terminal device identifier.

6. The method of claim 5, wherein, The perception model function identifier comprises one or more of: an identifier of a direct perception function or an identifier of an auxiliary perception function.

7. The method of claim 6, wherein, The direct perception function comprises one or more of: imaging, positioning, velocity estimation, target detection, or image segmentation; and / or The auxiliary perception function comprises determining an intermediate variable.

8. The method according to any one of claims 1 to 7, wherein The perception model input information satisfies at least one of: when the perception model is used for imaging, the perception model input information comprises one or more of: sampling information of an echo signal or a range image; when the perception model is used for target detection, the perception model input information comprises one or more of: sampling information of the echo signal or an imaging result; when the perception model is used for velocity estimation, the perception model input information comprises one or more of: channel quality information or a Doppler shift spectrum.

9. The method according to any one of claims 1 to 8, wherein, The perception model performance information satisfies at least one of: when the perception model is used for imaging, the perception model performance information comprises one or more of: an F1-score or an inverse angular distance or a ratio of intersection over union; when the perception model is used for target detection, the perception model performance information comprises one or more of: the F1-score, the inverse angular distance, or the ratio of intersection over union; when the perception model is used for velocity estimation, the perception model performance information comprises one or more of: accuracy or mean square error.

10. The method of any one of claims 1-9, wherein, The perception model available condition satisfies at least one of: when the perception model is used for imaging, the perception model available condition comprises one or more of: a condition that a receptive field and a resolution satisfy, a threshold of a time synchronization error, or a threshold of a position synchronization error; when the perception model is used for target detection, the perception model available condition comprises one or more of: a condition that a receptive field and a target size satisfy, the threshold of the time synchronization error, or the threshold of the position synchronization error; when the perception model is used for velocity estimation, the perception model available condition comprises one or more of: a condition that a receptive field and a preset time satisfy, or the threshold of the time synchronization error.

11. The method of claim 2 or 4, wherein, The first information comprises one or more of the following parameters: a required registration awareness model function, the awareness model identifier, the awareness model input information, the awareness model performance information, or the awareness model availability condition.

12. The method of claim 2, 4, or 11, wherein, The first information comprises an awareness model function identifier and a mapping relationship between an awareness model function and the awareness model function identifier.

13. The method of claim 2 or 4, wherein, The first information comprises awareness model architecture information.

14. The method of claim 13, wherein, The awareness model architecture information indicates at least one awareness model architecture corresponding to at least one awareness model function respectively.

15. The method of claim 13 or 14, wherein, The method further comprises: sending an awareness model requirement, the awareness model requirement being used to determine an awareness model architecture.

16. The method of claim 13 or 14, wherein, The method further comprises: receiving an awareness model requirement, the awareness model requirement being used to determine an awareness model architecture.

17. The method of claim 15 or 16, wherein, The awareness model requirement comprises one or more of the following: an awareness model function, the awareness model input information, or the awareness model availability condition.

18. The method of claim 2 or 4, wherein, The first information comprises a mapping relationship between an awareness model function and the awareness model architecture information.

19. A communications device, characterized by comprise units or modules for performing the method according to any one of claims 1-2, 5-15, 17-18, or comprise units or modules for performing the method according to any one of claims 3-14, 16-18.

20. A communications device, characterized by comprise a processor for executing computer programs or instructions to implement the method according to any one of claims 1-2, 5-15, 17-18, or implement the method according to any one of claims 3-14, 16-18.

21. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer programs or instructions, when the computer programs or instructions are executed by a communication device, the method according to any one of claims 1-2, 5-15, 17-18 is implemented, or the method according to any one of claims 3-14, 16-18 is implemented.

22. A computer program product, characterised in that, The computer program product contains computer programs or instructions, when the computer programs or instructions are executed by a computer, the method according to any one of claims 1-2, 5-15, 17-18 is implemented, or the method according to any one of claims 3-14, 16-18 is implemented.

23. A chip or chip system, characterized by The chip or chip system comprises a processor for executing the method according to any one of claims 1-2, 5-15, 17-18, or executing the method according to any one of claims 3-14, 16-18.

Images