Model selection method, first device, second device, apparatus, and system

Abstract

The present disclosure provides a model selection method, a first device, a second device, an apparatus, and a system. The method comprises: performing channel measurement on the basis of a first signal, and determining channel measurement data, wherein the first signal is a signal obtained when a sensing signal is received; and selecting, from among a plurality of second AI models, one second AI model matching environment information, wherein each second AI model is used for performing device positioning or beam management in different environments, and the environment information is environment information determined on the basis of the channel measurement data. The present disclosure improves the reliability of device positioning and beam management, and improves the availability of ISAC technology.

Description

Model selection method, first device, second device, apparatus and system TECHNICAL FIELD

[0001] The present disclosure relates to the field of communication, and in particular to a model selection method, a first device, a second device, an apparatus and a system. BACKGROUND

[0002] Integrated Sensing and Communication (ISAC) is one of the most important research directions in wireless communication. ISAC technology helps to achieve detection, positioning and tracking, environment reconstruction and target imaging, gesture and posture recognition and many other application services.

[0003] SUMMARY

[0004] In order to improve the usability of ISAC technology, the embodiments of the present disclosure provide a model selection method, a first device, a second device, an apparatus and a system.

[0005] According to a first aspect of the embodiments of the present disclosure, a model selection method is provided, which is performed by a first device, and includes:

[0006] performing channel measurement based on a first signal to determine channel measurement data; wherein the first signal is a signal obtained when receiving a sensing signal;

[0007] selecting one second artificial intelligence (AI) model matching environment information from a plurality of second AI models; wherein each second AI model is used for device positioning or beam management in different environments; and wherein the environment information is determined based on the channel measurement data.

[0008] According to a second aspect of the embodiments of the present disclosure, a model selection method is provided, which is performed by a second device, and includes:

[0009] selecting one second artificial intelligence (AI) model matching environment information from a plurality of second AI models; wherein each second AI model is used for device positioning or beam management in different environments; and wherein the environment information is determined based on channel measurement data.

[0010] sending indication information to the first device; wherein the indication information is used to indicate an identity of the second AI model matching the environment information.

[0011] According to a third aspect of the embodiments of the present disclosure, a first device is provided, which includes:

[0012] a processing module, configured to perform channel measurement based on the first signal to determine channel measurement data, wherein the first signal is a signal obtained when the sensing signal is received;

[0013] The processing module is further configured to select one second artificial intelligence (AI) model that matches the environment information from a plurality of second AI models, wherein each second AI model is used for device positioning or beam management in different environments, and wherein the environment information is determined based on the channel measurement data.

[0014] According to a fourth aspect of embodiments of the present disclosure, a second device is provided, comprising:

[0015] The processing module is further configured to select one second artificial intelligence (AI) model that matches the environment information from a plurality of second AI models, wherein each second AI model is used for device positioning or beam management in different environments, and wherein the environment information is determined based on the channel measurement data.

[0016] The transceiver module is configured to send indication information to the first device, wherein the indication information is used to indicate an identity of the second AI model that matches the environment information.

[0017] According to a fifth aspect of embodiments of the present disclosure, a communication apparatus is provided, the apparatus comprising a processor and a memory, the memory storing a computer program, and the processor executing the computer program stored in the memory to cause the apparatus to perform the method according to any one of the first aspect or the second aspect.

[0018] According to a sixth aspect of embodiments of the present disclosure, a communication system is provided, comprising:

[0019] a third device configured to send a sensing signal;

[0020] a first device configured to implement the method according to any one of the first aspect;

[0021] a second device configured to implement the method according to any one of the second aspect.

[0022] According to a seventh aspect of embodiments of the present disclosure, a storage medium is provided, the storage medium storing instructions that, when executed on a communication device, cause the communication device to perform the method according to any one of the first aspect or the second aspect.

[0023] According to an eighth aspect of embodiments of the present disclosure, a computer program product is provided, comprising a computer program configured to implement the method according to any one of the first aspect or the second aspect when executed by a processor.

[0024] In the embodiments of the present disclosure, the first device can select one second AI model matched with the environment information from a plurality of second AI models, each second AI model can perform device positioning or beam management in different environments, improve the reliability of device positioning and beam management, and improve the availability of ISAC technology.

[0025] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and are not limiting to the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0026] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments consistent with the present disclosure and serve to explain the principles of the present disclosure together with the specification.

[0027] FIG. 1A is one exemplary schematic diagram of an architecture of a communication system according to embodiments of the present disclosure.

[0028] FIG. 1B is an exemplary schematic diagram of a perception mode according to embodiments of the present disclosure.

[0029] FIG. 1C is an exemplary schematic diagram of four types of signals in a communication perception model according to embodiments of the present disclosure.

[0030] FIG. 1D is an exemplary schematic diagram of implementing device perception based on a perception algorithm according to embodiments of the present disclosure.

[0031] FIG. 1E is an exemplary schematic diagram of implementing device perception based on an AI model according to embodiments of the present disclosure.

[0032] FIG. 1F is an exemplary schematic diagram of multiple AI model selection and inference according to embodiments of the present disclosure.

[0033] FIG. 2A is one of the exemplary interactive schematic diagrams of a model selection method according to embodiments of the present disclosure.

[0034] FIG. 2B is another of the exemplary interactive schematic diagrams of a model selection method according to embodiments of the present disclosure.

[0035] FIG. 2C is an exemplary flow schematic diagram of a model training method according to embodiments of the present disclosure.

[0036] FIG. 3A is one of the exemplary flow schematic diagrams of a model selection method according to embodiments of the present disclosure.

[0037] FIG. 3B is another of the exemplary flow schematic diagrams of a model selection method according to embodiments of the present disclosure.

[0038] FIG. 3C is a third of the exemplary flow schematic diagrams of a model selection method according to embodiments of the present disclosure.

[0039] FIG. 3D is an example flow diagram four of a model selection method according to embodiments of the present disclosure.

[0040] FIG. 3E is an example flow diagram five of a model selection method according to embodiments of the present disclosure.

[0041] FIG. 4A is an example schematic diagram of AI model selection based on environment information according to embodiments of the present disclosure.

[0042] FIG. 4B is an example interaction schematic diagram of AI model selection based on environment information according to embodiments of the present disclosure.

[0043] FIG. 5A is an example block diagram of a first device according to embodiments of the present disclosure.

[0044] FIG. 5B is an example block diagram of a second device according to embodiments of the present disclosure.

[0045] FIG. 6A is an example interaction schematic diagram of a communication device according to embodiments of the present disclosure.

[0046] FIG. 6B is an example interaction schematic diagram of a chip according to embodiments of the present disclosure. DETAILED DESCRIPTION

[0047] The example embodiments will be described in detail herein with reference to the accompanying drawings. In the following description, the same numbers refer to the same or similar elements unless otherwise represented. The implementations described in the following example embodiments do not represent all implementations consistent with the present disclosure. Instead, they only represent examples of apparatuses and methods consistent with some aspects of the present disclosure, as detailed in the appended claims.

[0048] The embodiments of the present disclosure propose a model selection method, a first device, a second device, an apparatus and system, and a storage medium.

[0049] In a first aspect, the embodiments of the present disclosure propose a model selection method, which is performed by a first device, comprising: performing channel measurement based on a first signal to determine channel measurement data; wherein the first signal is a signal obtained when a sensing signal is received; selecting one second artificial intelligence (AI) model from a plurality of second AI models that matches environment information; wherein each second AI model is used for device positioning or beam management in different environments; and wherein the environment information is determined based on the channel measurement data.

[0050] In the above embodiment, the first device can select one of the plurality of second AI models for model inference, thereby improving the reliability of device positioning and beam management and improving the usability of the ISAC technology.

[0051] In some embodiments of the first aspect, the method further includes: inputting the channel measurement data into the first AI model to obtain the environment information output by the first AI model.

[0052] In the above embodiment, the environment information can be obtained through the first AI model, which is simple and easy to use.

[0053] In some embodiments of the first aspect, the method further includes: collecting sample channel measurement data corresponding to different environment information; inputting the sample channel measurement data into an initial AI model to obtain estimated environment information output by the initial AI model; determining a loss function based on the difference between the estimated environment information and the real environment information; training the initial AI model based on the loss function until the training is stopped when a stop condition is met, to obtain the first AI model.

[0054] In the above embodiment, the first AI model can be trained in the above manner, thereby improving the reliability of perceiving environment information.

[0055] In some embodiments of the first aspect, the stop condition includes at least one of the following: a number of training cycles is reached; the loss function is reduced to a fault tolerance range; the accuracy of the AI model after at least one round of training reaches a first value.

[0056] In the above embodiment, when the stop condition is met, the training can be stopped to obtain the first AI model, thereby improving the reliability and accuracy of training the first AI model.

[0057] In some embodiments of the first aspect, the method further includes: sending the environment information to a second device; receiving indication information sent by the second device; wherein the indication information is used to indicate the identity of the second AI model matched with the environment information.

[0058] In the above embodiment, the first device can send the environment information to the second device and receive the indication information sent by the second device, thereby selecting the second AI model, improving the reliability of device positioning and beam management, and improving the usability of the ISAC technology.

[0059] In some embodiments combined with the first aspect, in some embodiments, the method further includes: sending the channel measurement data to the second device; receiving indication information sent by the second device; wherein the indication information is used to indicate the identity of the second AI model matched with the environment information.

[0060] In the above embodiments, the first device can send the channel measurement data to the second device, so that the first AI model does not need to be deployed on the first device, saving the resources of the first device. In addition, the first device can receive the indication information sent by the second device, so as to select the second AI model, improve the reliability of device positioning and beam management, and improve the usability of ISAC technology.

[0061] In some embodiments combined with the first aspect, in some embodiments, the selecting, from a plurality of second artificial intelligence (AI) models, one second AI model matched with environment information includes: selecting, from the plurality of second AI models, the second AI model indicated by the indication information.

[0062] In the above embodiments, the first device can select the second AI model based on the indication information sent by the second device, improve the reliability of device positioning and beam management, and improve the usability of ISAC technology.

[0063] In some embodiments combined with the first aspect, in some embodiments, the selecting, from a plurality of second artificial intelligence (AI) models, one second AI model matched with environment information includes: selecting, from the plurality of second AI models, one second AI model matched with the environment information based on a correspondence between the second AI model and the environment information.

[0064] In the above embodiments, the first device can select the second AI model by itself, saving the signaling resources for interaction with the second device, improving the reliability of device positioning and beam management, and improving the usability of ISAC technology.

[0065] In some embodiments combined with the first aspect, in some embodiments, the environment information is used to indicate at least one of: an environment type; a scene type.

[0066] In the above embodiments, the environment information can indicate the environment type and / or the scene type, improving the reliability of device positioning and beam management, and improving the usability of ISAC technology.

[0067] In a second aspect, a model selection method is provided. The method is performed by a second device and includes: selecting, from a plurality of second artificial intelligence (AI) models, one second AI model that matches environment information; wherein each second AI model is configured to perform device positioning or beam management in a different environment; wherein the environment information is determined based on channel measurement data; and sending, to a first device, indication information; wherein the indication information is used to indicate an identity of the second AI model that matches the environment information.

[0068] In the above embodiments, the model selection is performed by the second device and indicated to the first device, thereby improving the reliability of device positioning and beam management and the usability of ISAC technology.

[0069] In some embodiments of the second aspect, the method further includes: receiving the channel measurement data sent by the first device; and inputting the channel measurement data into a first AI model to obtain the environment information output by the first AI model.

[0070] In some embodiments of the second aspect, the method further includes: collecting sample channel measurement data corresponding to different environment information; inputting the sample channel measurement data into an initial AI model to obtain estimated environment information output by the initial AI model; determining a loss function based on a difference between the estimated environment information and real environment information; and training the initial AI model based on the loss function until a stop condition is met, thereby obtaining the first AI model.

[0071] In some embodiments of the second aspect, the stop condition includes at least one of: a number of training cycles is reached; the loss function is reduced to a fault tolerance range; and an accuracy of an AI model after at least one round of training reaches a first value.

[0072] In some embodiments of the second aspect, the method further includes: receiving the environment information sent by the first device.

[0073] In some embodiments of the second aspect, the environment information is used to indicate at least one of: an environment type; and a scene type.

[0074] In a third aspect, an embodiment of the present disclosure provides a first device, comprising: a processing module configured to perform channel measurement based on a first signal to determine channel measurement data; wherein the first signal is a signal obtained when a sensing signal is received; the processing module is further configured to select one second artificial intelligence (AI) model matching environment information from a plurality of second AI models; wherein each second AI model is used for device positioning or beam management in different environments; and wherein the environment information is determined based on the channel measurement data.

[0075] In a fourth aspect, an embodiment of the present disclosure provides a second device, comprising: a processing module configured to select one second artificial intelligence (AI) model matching environment information from a plurality of second AI models; wherein each second AI model is used for device positioning or beam management in different environments; and wherein the environment information is determined based on channel measurement data; and a transceiver module configured to send indication information to the first device; wherein the indication information is used to indicate the identity of the second AI model matching the environment information.

[0076] In a fifth aspect, an embodiment of the present disclosure provides a communication apparatus, comprising a processor and a memory, wherein the memory stores a computer program, and the processor executes the computer program stored in the memory, so that the apparatus performs the method of any one of the first aspect or the second aspect.

[0077] In a sixth aspect, an embodiment of the present disclosure provides a communication system, comprising: a third device configured to send a sensing signal; a first device configured to implement the model selection method of any one of the first aspect; and a second device configured to implement the method of any one of the second aspect.

[0078] In a seventh aspect, an embodiment of the present disclosure provides a storage medium, which stores instructions, and when the instructions run on a communication device, the communication device performs the method of any one of the first aspect or the second aspect.

[0079] In an eighth aspect, an embodiment of the present disclosure provides a computer program product, comprising a computer program configured to implement the method of any one of the first aspect or the second aspect when executed by a processor.

[0080] In a ninth aspect, an embodiment of the present disclosure provides a chip or chip system. The chip or chip system comprises a processing circuit configured to perform the method described in the optional implementation of the first aspect or the second aspect.

[0081] It is understood that the first device, the second device, the communication system, the storage medium, the computer program product, the chip, or the chip system described above are all used to perform the methods proposed in the embodiments of this disclosure. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0082] The present invention is described in the embodiments described herein. In some embodiments, the terms model selection method, communication sensing method, model switching method, etc., can be used interchangeably; the terms model selection device, communication sensing device, model switching device, etc., can be used interchangeably; and the terms communication system, communication sensing system, model switching system, etc., can be used interchangeably.

[0083] This disclosure is not exhaustive, but merely illustrative of some embodiments, and is not intended to limit the scope of protection of this disclosure. Unless otherwise specified, each step in a particular embodiment can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, a solution after removing some steps in a particular embodiment can also be implemented as an independent embodiment, and the order of the steps in a particular embodiment can be arbitrarily interchanged. Furthermore, the optional implementation methods in a particular embodiment can be arbitrarily combined; moreover, the embodiments can be arbitrarily combined, for example, some or all steps of different embodiments can be arbitrarily combined, and a particular embodiment can be arbitrarily combined with the optional implementation methods of other embodiments.

[0084] In each of the disclosed embodiments, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of the embodiments are consistent and can be referenced by each other. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0085] The terminology used in the embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the scope of this disclosure.

[0086] In this embodiment of the disclosure, unless otherwise stated, elements expressed in the singular form, such as "a," "an," "the," "the," "the," "the," "the," "the," "this," etc., can mean "one and only one," or "one or more," "at least one," etc. For example, when using articles such as "a," "an," "the," etc. in translation, the noun following the article can be understood as either a singular expression or a plural expression.

[0087] In the embodiments disclosed herein, "multiple" refers to two or more.

[0088] In some embodiments, the terms "at least one of," "one or more of," "a plurality of," "multiple," and the like can be used interchangeably.

[0089] In some embodiments, the recitations "at least one of A, B," "A and / or B," "in one case A, in another case B," "in response to a case A, in response to a case B," and the like can include the following technical solutions according to the case: in some embodiments A (A is executed regardless of B); in some embodiments B (B is executed regardless of A); in some embodiments, A and B are selectively executed (A and B are selectively executed); in some embodiments, A and B (A and B are executed). When there are more branches such as A, B, C, and the like, the above is similar.

[0090] In some embodiments, the recitations "A or B" and the like can include the following technical solutions according to the case: in some embodiments A (A is executed regardless of B); in some embodiments B (B is executed regardless of A); in some embodiments, A and B are selectively executed (A and B are selectively executed). When there are more branches such as A, B, C, and the like, the above is similar.

[0091] In the embodiments of the present disclosure, the prefix words "first", "second", and the like are only used to distinguish different description objects, and do not constitute a limitation on the position, order, priority, quantity, or content of the description objects. The description of the description objects should refer to the description in the context of the claims or embodiments, and should not constitute an additional limitation because of the use of the prefix words. For example, the description object is "field", and the ordinal words before "field" in "first field" and "second field" do not limit the position or order between "fields", and "first" and "second" do not limit whether the "fields" modified thereby are in the same message or not, nor do they limit the order of "first field" and "second field". For another example, the description object is "level", and the ordinal words before "level" in "first level" and "second level" do not limit the priority between "levels". For another example, the quantity of the description object is not limited by the ordinal words, and can be one or more. For example, "first device", wherein the quantity of "device" can be one or more. In addition, the objects modified by different prefix words can be the same or different, for example, the description object is "device", and "first device" and "second device" can be the same device or different devices, and their types can be the same or different; for another example, the description object is "information", and "first information" and "second information" can be the same information or different information, and their contents can be the same or different.

[0092] In some embodiments, "comprising", "including", "to indicate", "carrying", can be interpreted as directly carrying A, or indirectly indicating A.

[0093] In some embodiments, the terms "in response to", "in response to determining", "in the case of", "when", "when", "if", "if" and the like can be replaced with each other.

[0094] In some embodiments, the terms "greater than", "greater than or equal to", "not less than", "more than", "more than or equal to", "not less than", "higher than", "higher than or equal to", "not lower than", "above", and the like can be replaced with each other, and the terms "less than", "less than or equal to", "not greater than", "less than", "less than or equal to", "not more than", "lower than", "lower than or equal to", "not higher than", "below", and the like can be replaced with each other.

[0095] In some embodiments, the device and the like can be interpreted as physical or virtual, and the name is not limited to the name described in the embodiments. The terms "device", "equipment", "device", "circuit", "network element", "node", "function", "unit", "section", "system", "network", "chip", "chip system", "entity", "subject" and the like can be replaced with each other.

[0096] In some embodiments, the data, information and the like can be obtained in accordance with the laws and regulations of the place.

[0097] In some embodiments, the data, information and the like can be obtained after obtaining the consent of the user.

[0098] In addition, each element, each row, or each column in the table of the embodiments of the present disclosure can be implemented as an independent embodiment, and any element, any row, any column combination can also be implemented as an independent embodiment.

[0099] FIG. 1A is a schematic diagram of the architecture of a communication system according to an embodiment of the present disclosure.

[0100] As shown in FIG. 1A, the communication system 100 includes but is not limited to a first device 101, a second device 102 and a third device 103.

[0101] In some embodiments, the first device 101 can be a sensing receiving end device, which can receive a sensing signal reflected, scattered, diffracted by a sensing target (a device that needs to be sensed) or a designated device (a device before the sensing target), thereby obtaining a first signal. Wherein, the first signal is the signal obtained when the first device 101 receives the sensing signal.

[0102] In one example, the first device 101 can include, but is not limited to, any one of the following: a network device, such as an access network device; a user equipment (UE); a vehicle.

[0103] In some embodiments, the second device 102 can be a device for model management, which can include, but is not limited to, a network device, such as an access network device, a core network device.

[0104] In some examples, the third device 103 can be a sensing transmitter device, which can transmit a sensing signal to a "sensing target" or a designated device, which reflects, scatters, diffracts the sensing signal, which is received by the first device 101 and obtained as a first signal.

[0105] Exemplarily, the third device 103 can include, but is not limited to, any one of the following devices: a vehicle; a user equipment (UE); an unmanned device; an Internet of Things (IoT) device; an Ambient Internet of Things (Ambient IoT) device.

[0106] Among them, the unmanned device can include, but is not limited to, a drone, an unmanned vehicle, etc.

[0107] Among them, the IoT device can include, but is not limited to, a smart wearable device, such as a smart watch, a smart bracelet, etc., a smart home device, such as a smart light bulb, a smart air conditioner, a smart rice cooker, etc., a smart meter, a smart water meter, etc.

[0108] Among them, the Ambient IoT device can include, but is not limited to, a device that transmits data and / or signaling after being triggered by other devices, such as terminals or network devices. It can be installed with a Radio Frequency Identification (RFID) tag, so it can be used as a reader by other devices, such as terminals or network devices, to perform tag inventory, data reporting, etc.

[0109] In some embodiments, the user equipment (UE) described above includes at least one of a mobile phone, a wearable device, an Internet of Things (IoT) device, a communication-capable automobile, a smart automobile, a tablet (Pad), a wireless-transmitting computer, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a wireless terminal device in self-driving, a wireless terminal device in remote medical surgery, a wireless terminal device in a smart grid, a wireless terminal device in transportation safety, a wireless terminal device in a smart city, a wireless terminal device in a smart home, and the like, but is not limited thereto.

[0110] In some embodiments, the access network device described above is at least one of a node or a device that accesses a terminal to a wireless network, and can include an evolved NodeB (eNB) in a 5G communication system, a next generation eNB (ng-eNB), a next generation NodeB (gNB), a node B (NB), a home node B (HNB), a home evolved node B (HeNB), a wireless backhaul device, a radio network controller (RNC), a base station controller (BSC), a base transceiver station (BTS), a base band unit (BBU), a mobile switching center, a base station in a 6G communication system, an Open RAN, a Cloud RAN, a base station in other communication systems, an access node in a Wi-Fi system, and the like, but is not limited thereto.

[0111] In some embodiments, the access network device can be composed of a central unit (CU) and a distributed unit (DU), where the CU can also be referred to as a control unit. The CU-DU structure can split the protocol layers of the access network device, with some of the protocol layers being controlled by the CU and the rest of the protocol layers or all of the protocol layers being distributed in the DUs and controlled by the CU, but is not limited thereto.

[0112] In some embodiments, the network device described above can also be a core network device. The core network device can be one device including one or more network elements, etc., or can be multiple devices or groups of devices. The network element can be virtual or physical. The core network includes at least one of an evolved packet core (EPC), a 5G core network (5GCN), and a next-generation core (NGC), for example.

[0113] In some embodiments, the technical solutions of the present disclosure can be applied to an Open RAN architecture. In this case, the interfaces between or within the access network devices described in the embodiments of the present disclosure can become internal interfaces of the Open RAN. The processes and information interactions between these internal interfaces can be implemented through software or programs.

[0114] It can be understood that the communication system described in the embodiments of the present disclosure is for more clearly illustrating the technical solutions of the present disclosure, and does not constitute a limitation on the technical solutions proposed in the embodiments of the present disclosure. It can be known by those skilled in the art that, as the system architecture evolves and new business scenarios appear, the technical solutions proposed in the embodiments of the present disclosure are also applicable to similar technical problems.

[0115] The following embodiments of the present disclosure can be applied to the communication system shown in FIG. 1A or part of the subject, but are not limited thereto. The subjects shown in FIG. 1A are examples. The communication system can include all or part of the subjects in FIG. 1A, or other subjects other than those in FIG. 1A. The number and form of the subjects are arbitrary. Each subject can be physical or virtual. The connection relationship between the subjects is an example. The subjects can not be connected or can be connected. The connection can be in any manner, can be direct or indirect, and can be wired or wireless.

[0116] The various embodiments of the present disclosure can be applied to Long Term Evolution (LTE), LTE-Advanced (LTE-A), LTE-Beyond (LTE-B), SUPER 3G, IMT-Advanced, 4th generation mobile communication system (4G), 5th generation mobile communication system (5G), 5G new radio (NR), Future Radio Access (FRA), New-Radio Access Technology (RAT), New Radio (NR), New radio access (NX), Future generation radio access (FX), Global System for Mobile communications (GSM (registered trademark)), CDMA2000, Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi (registered trademark)), IEEE 802.16 (WiMAX (registered trademark)), IEEE 802.20, Ultra-WideBand (UWB), Bluetooth (Bluetooth (registered trademark)), Public Land Mobile Network (PLMN) network, systems using other communication methods, next-generation systems expanded based on them, and the like. In addition, a plurality of systems can be combined (for example, a combination of LTE or LTE-A and 5G, and the like).

[0117] At present, the industry faces the communication and sensing integrated scene application, and based on the channel modeling related standards, a relatively complete channel modeling work of the communication and sensing system is carried out, which provides strong support for the evaluation of the communication and sensing technology scheme. The related research work proposes and designs various estimation algorithms of sensing target speed, distance, angle and other parameters for highway and other scenes, including multiple signal classification (MUSIC) algorithm, estimation of signal parameters using rotational invariance techniques (ESPRIT) algorithm, etc. These algorithms can achieve good target sensing effect in some communication and sensing integrated simulation scenes.

[0118] The application of artificial intelligence (AI) technology in the field of wireless communication is an important research direction of the academic and industrial circles, and is also one of the important research topics of 3GPP standardization. Machine learning (ML) and deep learning (DL) technologies have been widely applied in many fields. The powerful feature extraction and mapping relationship modeling capabilities of neural network models can provide new solutions to key problems in wireless communication systems, and have been widely researched and applied in high-precision terminal positioning, channel state information feedback, and beam management.

[0119] At the same time, the implementation of communication and sensing integration based on AI technology is also a potential application in future wireless communication systems. By using the powerful computing power of special devices such as graphics processing units (GPUs), AI sensing models are trained based on certain forms of data sets. Compared with traditional non-AI sensing algorithms, the AI sensing models can optimize the sensing accuracy, computational complexity and application scope, promote the deep application of communication and sensing integration technology, support more intelligent business applications and services, and thus help to promote the development of communication, sensing, calculation and intelligence integration.

[0120] In the embodiments of the present disclosure, the ISAC can include but is not limited to the following six sensing modes, as shown in FIG. 1B. Among them, mono-static means self-transmission and self-reception, that is, the same for the sending end and the receiving end; bi-static means different station transmission and reception, that is, different for the sending end and the receiving end.

[0121] Perception mode 1, self-transmission and self-reception, both the sending end and the receiving end are user equipment (User Equipment, UE), such as users (people). The sending end sends a perception signal to another UE (car), and receives the reflected echo of the other UE (car), thereby perceiving the distance, speed, angle, etc. of the UE itself (person).

[0122] Perception mode 2, self-transmission and self-reception, both the sending end and the receiving end are user equipment (User Equipment, UE), such as users (people). The sending end sends a perception signal to an access network device, such as a gNB, and receives a reflected echo through the gNB, thereby perceiving the distance, speed, angle, etc. of the UE itself (person).

[0123] Perception mode 3, different station transmission and reception, the sending end is a gNB, and the receiving end is a UE (car). For example, the gNB sends a perception signal, the UE (car) receives a signal forwarded through a user (person), and the UE (car) calculates a channel matrix based on the received signal to perceive the distance, speed, angle, etc. of a target such as a user (person) in the environment.

[0124] Perception mode 4, different station transmission and reception, the sending end is a UE (car), and the receiving end is a gNB. For example, the UE (car) sends a perception signal, the gNB receives a signal forwarded through a user (person), and the gNB calculates a channel matrix based on the received signal to perceive the distance, speed, angle, etc. of a target such as a user (person) in the environment.

[0125] Perception mode 5, different station transmission and reception, the sending end is a gNB#1, and the receiving end is a gNB#2. For example, the gNB#1 sends a perception signal, the gNB#2 receives a signal forwarded through a user (person), and the gNB#2 calculates a channel matrix based on the received signal to perceive the distance, speed, angle, etc. of a target such as a user (person) in the environment.

[0126] Perception mode 6, different station transmission and reception, the sending end is a UE#1 (car#1), and the receiving end is a UE#2 (car#2). For example, the UE#1 (car#1) sends a perception signal, the UE#2 (car#2) receives a signal forwarded through a user (person), and the UE#2 (car#2) calculates a channel matrix based on the received signal to perceive the distance, speed, angle, etc. of a target such as a user (person) in the environment.

[0127] It can be understood that the above perception targets can include: drones, people in indoor and outdoor scenes, cars on highways in outdoor scenes, automated guided vehicles in factories in indoor scenes, dangerous targets on roads or railways, etc. The integration of sensing and perception mainly involves the receiving end calculating and obtaining the distance, speed, angle, etc. of the perception target in the environment based on the received signal. The perception target can be the receiving end itself, or it can also be other objects.

[0128] In some embodiments, there are a sensing target, a base station, and an occlusion, which can be other UEs or clutter. Therefore, in the sensing channel model, four types of signals are mainly considered, which are a sensing Line Of Sight (LOS) path, a sensing Non-Line Of Sight (NLOS) path, a clutter LOS path, and a clutter NLOS path.

[0129] FIG. 1C is a schematic diagram of the sensing channel sensing LOS path, sensing NLOS path, clutter LOS, and clutter NLOS path in a user spontaneous sending and receiving scenario. Generally, the influence of channel noise also needs to be considered.

[0130] In FIG. 1C, the sensing LOS path: the path between the sensing signal reflection end and the sensing target is the sensing LOS path, and the path between the sensing target and the sensing signal receiving end is also the sensing LOS path.

[0131] The sensing NLOS path: the path between the sensing signal transmitting end and the sensing target that needs to pass through the clutter is the sensing NLOS path, and the path between the sensing target and the sensing signal receiving end that needs to pass through the clutter is also the sensing NLOS path.

[0132] The clutter LOS path: the path between the sensing signal transmitting end and the scattering cluster (for example, the clutter) in the environment is the clutter LOS path, and the path between the scattering cluster in the environment and the sensing signal receiving end is also the clutter LOS path.

[0133] The clutter NLOS path: the path between the sensing signal transmitting end and the scattering cluster (for example, the clutter) in the environment that still needs to pass through the clutter is the sensing NLOS path, and the path between the scattering cluster (for example, the clutter) in the environment and the sensing signal receiving end that still needs to pass through the clutter is also the sensing NLOS path.

[0134] In the embodiments of the present disclosure, the main information for sensing the target is the Channel State Information (CSI) matrix H of the receiving end, which is the superposition of the sensing target channel, the clutter channel, and the noise, and can reflect the channel environment in which the sensing signal is located and the state information of the sensing target in the environment. The dimension of the CSI matrix H is related to the specific parameter settings in the actual sensing scenario, and the information of the data on different dimensions can reflect different types of state information of the sensing target. In the embodiments of the present disclosure, one possible case is given:

[0135] The CSI matrix H usually contains three dimensions of subcarriers, Orthogonal Frequency Division Multiplexing (OFDM) symbols, and receiving antenna ports.

[0136] Specifically, the channel matrix H represents channel state information (CSI), which can be a complex matrix of MxSxP, where M is the number of OFDM symbols, S is the number of subcarriers, and P is the number of receive antenna ports. The element at the mth OFDM symbol, sth subcarrier, and pth receive antenna port position in the channel matrix H is a complex number, as shown in Equation 1, representing the amplitude and phase impact on the signal during its propagation from the transmitting antenna to the receiving antenna:

[0137] H m,s,p = Z m,s,p = a m,n,p + i x b m,s,p Equation 1

[0138] According to the received signal model, the phase shifts in the three dimensions of the channel matrix do not affect each other, and different types of perception results of the target can be obtained based on the phase changes in the three dimensions, i.e., the phase difference caused by the time delay between subcarriers can be used to estimate the distance of the target, the phase difference caused by the Doppler effect between OFDM symbols can be used to estimate the speed of the target, and the phase changes between multiple antenna ports can be used to estimate the angle of the target.

[0139] In some embodiments, the perception results of the three parameters of the target, such as distance, speed, and angle, can be obtained based on the CSI matrix through a perception algorithm, as shown in FIG. ID.

[0140] Using the strong learning and modeling capabilities of neural network models, an AI perception model can be obtained by training a certain amount of data, which can complete the estimation of the number of perception targets and the perception of the distance, speed, angle, and other information of each target in a multi-target perception scenario. For example, the channel state information matrix H is directly used as the input data of the AI perception model, and an AI perception model is trained, which simultaneously outputs the distance speed angle of the target in three types of perception results, as shown in FIG. IE.

[0141] The main application process of the AI model in the communication system is as follows: after the training data is used to complete the training of the AI neural network model (Model Training), the trained AI model is deployed on the actual device in the communication system, the model inference (Model Inference) is performed to obtain the results, and then the corresponding tasks in the actual system are completed.

[0142] Since the storage and computing capabilities of the related devices in the actual communication system are limited, when the AI scheme is used to solve the problems in the wireless communication system, the size and computational complexity of the neural network model used are also limited. In this case, the generalization problem of the neural network model is more serious.

[0143] The generalization capability of a neural network model refers to the performance of the model when processing unknown data different from training data. Generally, the model can better process data conforming to the same or similar distribution as the training data set, and when the model needs to process data conforming to a feature distribution significantly different from the training data set, the model performance will be significantly reduced. The channel scenarios or channel conditions in the actual communication system have diversity, and under different channel scenarios or channel conditions, the feature distribution of the channel measurement data used to complete terminal positioning may have large differences, so the AI model applied to a certain channel scenario or channel condition is difficult to maintain good model performance under other conditions.

[0144] In order to improve the generalization of neural network models in actual system applications, the following three solutions can be used, but are not limited to:

[0145] Method 1) Use only one AI model under different channel scenarios and / or channel conditions.

[0146] This method usually needs to use a mixed data set under different channel scenarios and / or channel conditions to complete AI model training, so that the model learns the data distribution under different channel conditions. It uses one model to process data under different conditions, and the model application process is simple, and the overall model size is low. However, if the data distribution under different channel scenarios and / or channel conditions is significantly different, using only one AI model cannot guarantee the model performance under various conditions.

[0147] Method 2) Use one model and perform model parameter fine-tuning or updating.

[0148] After deploying the pre-trained model on the device, the model parameter fine-tuning or updating is performed using the data under the actual condition in the model application process to realize the learning of the model to the current data features. The overall model size is low and can achieve good model performance. However, the data collection in the actual system, the overhead of model parameter fine-tuning or updating is large, which causes large time delay and device power consumption.

[0149] Method 3) Use multiple models and implement model switching.

[0150] To ensure the model performance under different conditions, different models are trained and deployed under different channel scenarios or channel conditions. Before model inference application, AI model selection is first completed, the AI model suitable for the current condition is selected and activated. It can achieve good model performance under various conditions, but the overall model size is high and model selection needs to be completed.

[0151] For method 3) for solving the AI model generalization problem, multiple models are used and model switching is implemented. To ensure high accuracy of model inference results, a method of training and storing multiple sets of network models for different channel scenarios or channel conditions is usually used. The appropriate AI model is used under different channel scenarios or channel conditions, and the AI model selection problem needs to be considered, as shown in FIG. 1F.

[0152] To achieve more accurate AI model selection, the present disclosure provides the following model selection method, first device, second device, apparatus and system.

[0153] FIG. 2A is an interaction diagram of a model selection method according to an embodiment of the present disclosure. As shown in FIG. 2A, the present disclosure relates to a model selection method, and the method includes:

[0154] In step S2101, the first device 101 receives a first signal.

[0155] In some embodiments, the first device 101 can be a perception receiving end device.

[0156] In some embodiments, the first device 101 can include but is not limited to a terminal, a network side device, an Internet of Things device, an environmental Internet of Things device, etc.

[0157] In some embodiments, the third device 103 (perception sending end device) sends a perception signal (or called a perception reference signal) to a perception target or a designated device, the perception target or the designated device reflects, scatters, diffracts the perception signal, and the first device 101 can obtain the first signal when receiving the perception signal.

[0158] In step S2102, the first device 101 performs channel measurement based on the first signal, and determines channel measurement data.

[0159] In some embodiments, the channel measurement data can include but is not limited to a channel state information matrix H. The CSI matrix H is the superposition of the perception target channel, the clutter channel and the noise, and can reflect the channel environment in which the perception signal is located and the state information of the perception target in the environment.

[0160] In step S2103, the first device 101 determines environmental information based on the channel measurement data.

[0161] In some embodiments, the environmental information can be used to indicate at least one of the following: an environment type; a scene type.

[0162] Exemplarily, the environmental information can be used to indicate indoor, outdoor, environmental complexity, etc.

[0163] In some embodiments, the first device 101 can input the channel measurement data into a first AI model, and obtain the environment information output by the first AI model.

[0164] The first AI model can be used for perception of the environment information.

[0165] The training process of the first AI model will be described in subsequent embodiments, which will not be described here.

[0166] In step S2104, the first device 101 selects one of the second AI models that matches the environment information from the plurality of second AI models.

[0167] In some embodiments, each second AI model can be used for device positioning or beam management in different environments, which is not limited by the present disclosure.

[0168] It can be understood that the second AI model is not limited to implementing device positioning or beam management, and can also be used to implement other functions.

[0169] When multiple second AI models for implementing the same function are deployed on the first device 101, the selection of the second AI model can be performed by the first device 101.

[0170] In one example, the first device 101 can select one second AI model that matches the current environment information from the plurality of second AI models based on the correspondence between the second AI model and the environment information. The correspondence can be that one second AI model corresponds to one environment or multiple environments, which is not limited by the present disclosure.

[0171] In some embodiments, the names of information and the like are not limited to the names described in the embodiments, and the terms such as "information", "message", "signal", "signaling", "report", "configuration", "indication", "instruction", "command", "channel", "parameter", "domain", "field", "symbol", "symbol", "codebook", "codeword", "code point", "bit", "data", "program", "chip", and the like can be replaced with each other.

[0172] In some embodiments, the terms of “sending”, “reflecting”, “reporting”, “issuing”, “transmitting”, “bidirectional transmission”, “sending and / or receiving” and the like can be replaced by each other.

[0173] In some embodiments, the terms of “acquiring”, “obtaining”, “getting”, “receiving”, “transmitting”, “bidirectional transmission”, “sending and / or receiving” can be replaced by each other, which can be interpreted as receiving from other subjects, acquiring from protocols, acquiring from higher layers, processing to get, autonomously implementing and the like.

[0174] In some embodiments, the terms of “certain”, “preseted”, “preset”, “set”, “indicated”, “certain”, “arbitrary”, “first”, “designated” and the like can be replaced by each other, “certain A”, “preset A”, “preset A”, “set A”, “indicated A”, “certain A”, “arbitrary A”, “first A” can be interpreted as A specified in advance in protocols and the like, A obtained by setting, configuration, or indication and the like, A specific, certain, arbitrary, or first A and the like, but not limited thereto.

[0175] In some embodiments, the device perception method related to the embodiments of the present disclosure can include at least one of steps S2101-S2104. For example, step S2101 can be implemented as an independent embodiment, step S2102 can be implemented as an independent embodiment, steps S2101+S2102 can be implemented as an independent embodiment, step S2103 can be implemented as an independent embodiment, step S2104 can be implemented as an independent embodiment, steps S2103+S2104 can be implemented as an independent embodiment, steps S2101-S2104 can be implemented as an independent embodiment, but not limited thereto.

[0176] In some embodiments, step S2103 is optional, and one or more of these steps can be omitted or replaced in different embodiments. For example, when the environment information is determined by other execution subjects such as the second device 102, step S2103 can not be performed.

[0177] In some embodiments, step S2104 is optional, and one or more of these steps can be omitted or replaced in different embodiments. For example, when the model selection is performed by other execution subjects such as the second device 102, step S2104 can not be performed.

[0178] In some embodiments, steps S2101 to S2104 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0179] In some embodiments, the execution order of steps S2101-S2104 is not limited.

[0180] In the above embodiments, the first device can select one second AI model matching the environment information after determining the environment information, thereby improving the reliability of device positioning and beam management and the usability of ISAC technology. In addition, the perceived environment information can also be used to improve the reliability of other related technologies.

[0181] FIG. 2B is an interaction diagram of a model selection method according to an embodiment of the present disclosure. As shown in FIG. 2B, the present embodiment relates to a model selection method, and the method includes:

[0182] In step S2201, the first device 101 receives a first signal.

[0183] In some embodiments, the implementation of step S2201 is similar to that of step S2101, which will not be repeated here.

[0184] In step S2202, the first device 101 performs channel measurement based on the first signal to determine channel measurement data.

[0185] In some embodiments, the implementation of step S2202 is similar to that of step S2102, which will not be repeated here.

[0186] In step S2203a, the first device 101 determines environment information based on the channel measurement data.

[0187] In some embodiments, the implementation of step S2203 is similar to that of step S2103, which will not be repeated here.

[0188] In step S2203b, the first device 101 sends the environment information to the second device 102.

[0189] In some embodiments, the second device 102 can be a device for model selection or model management, including but not limited to an access network device and a core network element.

[0190] In some embodiments, the second device 102 receives the environment information.

[0191] In step S2204a, the first device 101 sends the channel measurement data to the second device 102.

[0192] In some embodiments, the second device 102 receives the channel measurement data.

[0193] In step S2204b, the second device 102 determines the environment information based on the channel measurement data.

[0194] In some embodiments, the second device 102 can input the channel measurement data into the first AI model, and obtain the environment information output by the first AI model.

[0195] The first AI model can be used for perception of the environment information.

[0196] The training process of the first AI model will be described in subsequent embodiments, which will not be described here.

[0197] In some embodiments, steps S2203a-S2203b and steps S2204a-S2204b can be executed alternatively. For example, the first AI model is deployed on the first device 101, considering that reporting channel measurement data requires occupying more resources, the first device 101 can determine the environment information by the first AI model, and then send the environment information to the second device 102, at this time, only steps S2203a-S2203b can be executed. For another example, the first AI model is not deployed on the first device 101, the first device 101 can send the channel measurement data to the second device 102, and the second device 102 determines the environment information by the first AI model, at this time, only steps S2204a-S2204b can be executed.

[0198] In step S2205, the second device 102 selects one of the second AI models that matches the environment information from the plurality of second AI models.

[0199] In some embodiments, each second AI model can be used for device positioning or beam management in different environments, which is not limited by the present disclosure.

[0200] It can be understood that the second AI model is not limited to implementing device positioning or beam management, and can also be used to implement other functions.

[0201] When the first device 101 is deployed with multiple second AI models for implementing the same function, the second device 102 can select a second AI model for the first device 101.

[0202] In one example, the second device 102 can select one of the second AI models that matches the current environment information from the plurality of second AI models based on a correspondence between the second AI models and the environment information. The correspondence can be that one second AI model corresponds to one environment or multiple environments, which is not limited by the present disclosure.

[0203] In step S2206, the second device 102 sends the indication information to the first device 101.

[0204] In some embodiments, the indication information can be used to indicate the identity of the second AI model matched with the environment information.

[0205] In some embodiments, the first device 101 receives the indication information.

[0206] In step S2207, the first device 101 selects the second AI model indicated by the indication information.

[0207] In some embodiments, a plurality of second AI models are deployed on the first device 101, and the first device 101 can select one second AI model based on the indication information.

[0208] In some embodiments, steps S2201 to S2207 are optional, and one or more of the steps can be omitted or replaced in different embodiments.

[0209] In some embodiments, the execution order of steps S2201 to S2207 is not limited.

[0210] In the above embodiments, the first device can send the channel measurement data or the environment information to the second device, and the second device can select one second AI model matched with the environment information from a plurality of second AI models, thereby improving the reliability of device positioning and beam management, improving the usability of ISAC technology, and further improving the reliability of other related technologies by using the perceived environment information.

[0211] FIG. 2C is a flow diagram of a model training method according to an embodiment of the present disclosure. As shown in FIG. 2C, the present disclosure relates to a model training method, which can be performed by the first device 101 or the second device 102. The method includes:

[0212] In step S2301, sample channel measurement data corresponding to different environment information is collected.

[0213] In some embodiments, the first device 101 or the second device 102 can collect corresponding sample channel measurement data in different environments, such as indoor, outdoor, and the like. The sample channel measurement data can include, but is not limited to, sample CSI matrices.

[0214] In step S2302, the sample channel measurement data is input into an initial AI model, and estimated environment information output by the initial AI model is obtained.

[0215] In some embodiments, the initial AI model can take a residual network (resnet), a Visual Geometry Group (VGG) network, or the like as a backbone network, and can include, but is not limited to, at least one of the following network layers: an input layer; a convolutional layer; a pooling layer; an activation function layer; a connection layer; and an output layer. The disclosure does not limit the architecture of the initial AI model.

[0216] The first device 101 or the second device 102 can input the sample channel measurement data into the initial AI model described above, and obtain estimated environment information output by the initial AI model.

[0217] In step S2303, a loss function is determined based on a difference between the estimated environment information and the real environment information.

[0218] In some embodiments, the first device 101 or the second device 102 can determine the real environment information in advance, calculate the difference between the estimated environment information and the real environment information, and thus determine the loss function.

[0219] In step S2304, the initial AI model is trained based on the loss function, and the training is stopped when a stop condition is met, and the first AI model is obtained.

[0220] In some embodiments, the stop condition can include, but is not limited to, at least one of the following:

[0221] The number of training cycles is reached;

[0222] The loss function is reduced to a fault-tolerant range;

[0223] The accuracy of the AI model after at least one round of training reaches a first value.

[0224] The first value can be a preset accuracy threshold. When the accuracy of the AI model after at least one round of training reaches the first value, it can be determined that the model accuracy meets or reaches the preset accuracy threshold, at which point the training can be stopped, and the obtained first AI model can be used for environment information perception.

[0225] In some embodiments, when the above-mentioned stop condition is met, the training is stopped, and thus the first AI model is obtained, which can be used for environment information perception.

[0226] In some embodiments, steps S2301 to S2304 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0227] In some embodiments, the execution order of steps S2301 to S2304 is not limited.

[0228] In the above embodiments, the first device or the second device can train the first AI model in the above manner, and the first AI model can be used for environment information perception, improving the reliability of determining the device environment and further improving the usability of the ISAC technology.

[0229] FIG. 3A is a flow diagram of a model selection method according to an embodiment of the present disclosure. As shown in FIG. 3A, the embodiments of the present disclosure relate to a device perception method, which can be performed by the first device 101, and the method comprises:

[0230] In step S3101, a first signal is obtained.

[0231] In some embodiments, the first device 101 obtains the first signal when receiving the perception signal.

[0232] In some embodiments, the first device 101 obtains the first signal specified by a protocol.

[0233] In some embodiments, the first device 101 obtains the first signal from an upper layer.

[0234] In some embodiments, the first device 101 processes to obtain the first signal.

[0235] In some embodiments, step S3101 is omitted, and the first device 101 autonomously implements the function indicated by the first signal, or the first device 101 obtains the first signal based on a pre-defined rule or protocol agreement, or the above function is default or default.

[0236] In some embodiments, the optional implementation of step S3101 can refer to the optional implementation of step S2101 in FIG. 2A and other associated parts in the embodiments related to FIG. 2A, which will not be repeated here.

[0237] In step S3102, channel measurement data is determined.

[0238] In some embodiments, the optional implementation of step S3102 can refer to the optional implementation of step S2102 in FIG. 2A and other associated parts in the embodiments related to FIG. 2A, which will not be repeated here.

[0239] In step S3103, environment information is determined.

[0240] In some embodiments, the optional implementation of step S3103 can refer to the optional implementation of step S2103 in FIG. 2A and other associated parts in the embodiments related to FIG. 2A, which will not be repeated here.

[0241] Step S3104. Among the plurality of second AI models, one second AI model that matches the environment information is selected.

[0242] In some embodiments, the optional implementation of step S3104 can refer to the optional implementation of step S2104 in FIG. 2A and other associated parts in the embodiments involved in FIG. 2A, which are not described here again.

[0243] In some embodiments, steps S3101 to S3104 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0244] In some embodiments, the execution order of steps S3101 to S3104 is not limited.

[0245] In the above embodiments, the first device can select one second AI model that matches the environment information after determining the environment information, thereby improving the reliability of device positioning and beam management and improving the usability of ISAC technology.

[0246] FIG. 3B is a flow diagram of a model selection method according to an embodiment of the present disclosure. As shown in FIG. 3B, the embodiments of the present disclosure relate to a device perception method, which can be performed by the first device 101, and the method comprises:

[0247] Step S3201. A first signal is obtained.

[0248] In some embodiments, the first device 101 obtains the first signal when receiving the perception signal.

[0249] In some embodiments, the first device 101 obtains the first signal specified by a protocol.

[0250] In some embodiments, the first device 101 obtains the first signal from an upper layer.

[0251] In some embodiments, the first device 101 processes to obtain the first signal.

[0252] In some embodiments, step S3101 is omitted, and the first device 101 autonomously implements the function indicated by the first signal, or the first device 101 obtains the first signal based on a predefined rule or protocol agreement, or the above function is default or default.

[0253] In some embodiments, the optional implementation of step S3201 can refer to the optional implementation of step S2201 in FIG. 2B and other associated parts in the embodiments involved in FIG. 2B, which are not described here again.

[0254] Step S3202. Channel measurement data is determined.

[0255] In some embodiments, the optional implementation of step S3202 can refer to the optional implementation of step S2202 in FIG. 2B and other associated parts in the embodiments involved in FIG. 2B, which will not be repeated here.

[0256] Step S3203a, determining environment information.

[0257] In some embodiments, the optional implementation of step S3203a can refer to the optional implementation of step S2203a in FIG. 2B and other associated parts in the embodiments involved in FIG. 2B, which will not be repeated here.

[0258] Step S3203b, sending environment information.

[0259] In some embodiments, the first device 101 sends the environment information to the second device 102.

[0260] In some embodiments, the second device 102 receives the environment information.

[0261] In some embodiments, the optional implementation of step S3203b can refer to the optional implementation of step S2203b in FIG. 2B and other associated parts in the embodiments involved in FIG. 2B, which will not be repeated here.

[0262] Step S3204, sending channel measurement data.

[0263] In some embodiments, the first device 101 sends the channel measurement data to the second device 102.

[0264] In some embodiments, the second device 102 receives the channel measurement data.

[0265] In some embodiments, the optional implementation of step S3204 can refer to the optional implementation of step S2204a in FIG. 2B and other associated parts in the embodiments involved in FIG. 2B, which will not be repeated here.

[0266] Step S3205, obtaining indication information.

[0267] In some embodiments, the indication information can be used to indicate the identity of the second AI model matched with the environment information.

[0268] In some embodiments, the first device 101 can obtain the indication information from the second device 102, but is not limited thereto, and can also receive indication information sent by other subjects.

[0269] In some embodiments, the first device 101 obtains the indication information specified by the protocol.

[0270] In some embodiments, the first device 101 obtains the indication information from upper layer(s).

[0271] In some embodiments, the first device 101 processes to obtain the indication information.

[0272] In some embodiments, step S3205 is omitted, and the first device 101 autonomously implements the function indicated by the indication information, or the first device 101 obtains the indication information based on a pre-defined rule or protocol agreement, or the above function is default or default.

[0273] In some embodiments, the optional implementation of step S3205 can refer to the optional implementation of step S2206 in FIG. 2B and other associated parts in the embodiments involved in FIG. 2B, which will not be repeated here.

[0274] Step S3206, selecting the second AI model indicated by the indication information.

[0275] In some embodiments, the optional implementation of step S3206 can refer to the optional implementation of step S2207 in FIG. 2B and other associated parts in the embodiments involved in FIG. 2B, which will not be repeated here.

[0276] In some embodiments, steps S3201 to S3206 are optional, and one or more of the steps can be omitted or replaced in different embodiments.

[0277] In some embodiments, the execution order of steps S3201 to S3206 is not limited.

[0278] In the above embodiments, the first device can send channel measurement data or environment information to the second device, and the second device selects one second AI model matching the environment information from multiple second AI models, which improves the reliability of device positioning and beam management and improves the usability of ISAC technology.

[0279] FIG. 3C is a flow diagram of a model selection method according to an embodiment of the present disclosure. As shown in FIG. 3C, the embodiment of the present disclosure relates to a device perception method, which can be performed by the first device 101, and the method comprises:

[0280] Step S3301, determining channel measurement data.

[0281] In some embodiments, the optional implementation of step S3301 can refer to the optional implementation of step S2102 in FIG. 2A and other associated parts in the embodiments involved in FIG. 2A, which will not be repeated here.

[0282] Step S3302: Select one of the second AI models that matches the environmental information from among the multiple second AI models.

[0283] In some embodiments, optional implementations of step S3302 can be found in optional implementations of step S2104 in FIG2A and other related parts in the embodiments involved in FIG2A, which will not be repeated here.

[0284] In some embodiments, optional implementations of step S3302 can be found in optional implementations of step S2207 in FIG2B and other related parts in the embodiments involved in FIG2B, which will not be repeated here.

[0285] In some embodiments, steps S3301 to S3302 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0286] In some embodiments, the execution order of steps S3301 to S3302 is not limited.

[0287] The above embodiments improve the reliability of device positioning and beam management, and enhance the usability of ISAC technology.

[0288] Figure 3D is a flowchart illustrating a model selection method according to an embodiment of the present disclosure. As shown in Figure 3D, this disclosure relates to a device sensing method, which can be executed by a second device 102, and includes:

[0289] Step S3401: Obtain environmental information.

[0290] In some embodiments, the second device 102 may obtain environmental information from the first device 101, but is not limited thereto, and may also receive environmental information sent by other entities.

[0291] In some embodiments, the second device 102 acquires environmental information as defined by the protocol.

[0292] In some embodiments, the second device 102 obtains environmental information from upper layer(s).

[0293] In some embodiments, the second device 102 processes environmental information to obtain environmental information.

[0294] In some embodiments, step S3401 is omitted, the second device 102 autonomously implements the function indicated by the environmental information, or the second device 102 obtains the environmental information based on predefined rules or protocol agreements, or the above function is a default or default setting.

[0295] In some embodiments, the optional implementation of step S3401 can refer to the optional implementation of step S2203b in FIG. 2B and other associated parts in the embodiments involved in FIG. 2B, which will not be repeated here.

[0296] Step S3402: Obtain channel measurement data.

[0297] In some embodiments, the second device 102 can obtain the channel measurement data from the first device 101, but is not limited thereto, and can also receive channel measurement data sent by other subjects.

[0298] In some embodiments, the second device 102 obtains channel measurement data specified by a protocol.

[0299] In some embodiments, the second device 102 obtains channel measurement data from upper layer(s).

[0300] In some embodiments, the second device 102 processes to obtain channel measurement data.

[0301] In some embodiments, step S3402 is omitted, and the second device 102 autonomously implements the function indicated by the channel measurement data, or the second device 102 obtains the channel measurement data based on a pre-defined rule or protocol agreement, or the above function is default or default.

[0302] In some embodiments, the optional implementation of step S3402 can refer to the optional implementation of step S2204a in FIG. 2B and other associated parts in the embodiments involved in FIG. 2B, which will not be repeated here.

[0303] Step S3403: Determine environment information.

[0304] In some embodiments, the optional implementation of step S3403 can refer to the optional implementation of step S2204b in FIG. 2B and other associated parts in the embodiments involved in FIG. 2B, which will not be repeated here.

[0305] Step S3404: Among a plurality of second AI models, select one second AI model matching the environment information.

[0306] In some embodiments, the optional implementation of step S3404 can refer to the optional implementation of step S2205 in FIG. 2B and other associated parts in the embodiments involved in FIG. 2B, which will not be repeated here.

[0307] Step S3405: Send indication information.

[0308] In some embodiments, the indication information can be used to indicate the identity of the second AI model matching the environment information.

[0309] In some embodiments, the second device 102 sends the indication information to the first device 101.

[0310] In some embodiments, the first device 101 receives the indication information.

[0311] In some embodiments, the optional implementation of step S3404 can refer to the optional implementation of step S2205 in FIG. 2B and other associated parts in the embodiments involved in FIG. 2B, which will not be described here.

[0312] In some embodiments, steps S3401 to S3404 are optional, and one or more of these steps can be omitted or replaced in different embodiments.

[0313] In some embodiments, the execution order of steps S3401 to S3404 is not limited.

[0314] In the above embodiments, the second device can obtain channel measurement data or environment information from the first device, so that the second device selects one second AI model matching the environment information from multiple second AI models, thereby improving the reliability of device positioning and beam management and improving the usability of ISAC technology.

[0315] FIG. 3E is a flow diagram of a model selection method according to an embodiment of the present disclosure. As shown in FIG. 3E, the present disclosure relates to a device perception method, which can be performed by the second device 102, and the method comprises:

[0316] Step S3501, selecting one second AI model matching the environment information from multiple second AI models.

[0317] In some embodiments, the optional implementation of step S3501 can refer to the optional implementation of step S2205 in FIG. 2B and other associated parts in the embodiments involved in FIG. 2B, which will not be described here.

[0318] Step S3502, sending the indication information.

[0319] In some embodiments, the indication information can be used to indicate the identity of the second AI model matching the environment information.

[0320] In some embodiments, the second device 102 sends the indication information to the first device 101.

[0321] In some embodiments, the first device 101 receives the indication information.

[0322] In some embodiments, the optional implementation of step S3502 can refer to the optional implementation of step S2205 in FIG. 2B and other associated parts in the embodiments related to FIG. 2B, which are not described herein again.

[0323] In some embodiments, steps S3501-S3502 are optional, and one or more of the steps can be omitted or replaced in different embodiments.

[0324] In some embodiments, the execution order of steps S3501-S3502 is not limited.

[0325] In the above embodiments, the second device can select one second AI model that matches the environment information from multiple second AI models, improving the reliability of device positioning and beam management and the usability of ISAC technology.

[0326] The above process is further illustrated as follows.

[0327] In the communication and sensing integrated application scenario, non-AI sensing algorithms or AI-based neural network models can be used to complete the sensing of the distance, speed, angle, etc. of the device itself or other targets in the environment of the device. At the same time, when AI models are used to achieve high-precision positioning, beam management, etc., the models have certain generalization problems, and in some cases multiple models need to be deployed and selected to achieve stable AI model application performance.

[0328] In the embodiments of the present disclosure, for AI-based wireless communication physical layer application technology, to ensure good model application performance, AI model selection needs to be completed. Before AI model inference, an AI model that matches the input data needs to be selected first. In the communication and sensing integrated scenario, the device can obtain its own environment, state, etc. information using sensing technology. Therefore, based on the sensing information, the AI model selection can be completed with high accuracy, and the corresponding AI model can be activated, and then the AI model suitable for the current channel scenario or channel condition can be used to obtain the inference result to ensure high model application performance.

[0329] In the communication and sensing integrated application scenario, the sensing signal sending end sends the sensing signal, and the sensing signal is received by the sensing signal receiving end after the reflection, refraction, scattering, etc. of the vehicle, base station or other objects. Based on the received signal, channel measurement data (such as channel state information matrix) for sensing can be obtained, which can reflect the current environment, channel condition, etc. of the device. Based on the channel measurement data for sensing, non-AI sensing algorithms or AI sensing models can be used to obtain the distance, speed, angle, etc. of the device itself.

[0330] For the application cases of AI technology in physical layer communication such as high-precision positioning and beam management, different AI models need to be applied to obtain results for different application scenarios or channel conditions. In the case of deploying multiple AI models and considering model selection, the present disclosure proposes an AI model selection scheme based on perception information, as shown in FIG. 4A.

[0331] (1) Obtain environment information based on AI environment perception model

[0332] Non-AI perception algorithms usually support calculating and obtaining distance, speed, angle, etc. of the perception target based on channel measurement data (such as channel state information matrix) for perception, but do not support obtaining environment information of the device based on channel measurement data for perception, including channel scene type such as indoor, outdoor, or environment complexity, etc.

[0333] The embodiment of the present disclosure proposes to realize environment information perception based on AI neural network model, which inputs channel measurement data for perception and outputs environment or channel information of the device. Data collection under different channel scene types and environment complexity settings obtains channel measurement data and its corresponding perception environment information label, which constitutes the model training data set. The AI environment information perception model is obtained by training using the data set. The model can learn the mapping relationship between channel measurement data and environment information in the model training process, and then realize the acquisition of environment information based on real-time input channel measurement data in actual application.

[0334] (2) Complete AI model selection based on environment information

[0335] After the first device (perception receiving end device) obtains environment information based on the first AI model, it can select one of the following two ways to complete the second model selection and model reasoning according to actual situation or actual system requirement, for example, as shown in FIG. 4B, including the following steps:

[0336] Step S4201, the first device obtains perception measurement data.

[0337] Step S4202, the first device reports the perception measurement data to the second device.

[0338] Step S4203, the first device perceives AI environment information.

[0339] Step S4203', the second device perceives AI environment information based on the perception measurement data.

[0340] Mode one, after step S4203 is executed, step S4207 is executed directly, the first device directly completes the second model selection and other steps, and uses the corresponding second model to perform model reasoning to obtain the result;

[0341] In the second mode, after step S4203 is performed, step S4204 is performed, and the first device reports the sensing environment information to the second device.

[0342] In step S4205, the second device selects a model based on the environment information determined in step S4203' or step S4204.

[0343] In step S4206, the second device indicates the model selection result.

[0344] After step S4207 or step S4206 is performed, step S4208 can be performed, and the first device uses the corresponding second model to perform model inference to obtain a result.

[0345] In the above embodiments, for the model selection problem existing in the actual system application process of the AI model, a training and application method of an AI environment information sensing model is proposed. Based on the channel measurement data for sensing, the AI environment information sensing model is used to obtain the environment information of the device, and then the AI model selection is assisted to complete. The AI model matched with the environment of the device is used to complete inference to obtain a result, so as to guarantee the better performance in the AI model application process such as high-precision positioning and beam management.

[0346] In the embodiments of the present disclosure, part or all of the steps and optional implementation manners thereof can be combined with part or all of the steps in other embodiments, or can be combined with optional implementation manners of other embodiments.

[0347] The embodiments of the present disclosure also propose a device for implementing any of the above methods. For example, a device is proposed, which includes units or modules for implementing each step performed by the first device in any of the above methods. For another example, another device is proposed, which includes units or modules for implementing each step performed by the second device in any of the above methods.

[0348] It should be understood that the division of each unit or module in the above apparatus is only a logical function division, and all or part of them can be integrated into a physical entity or physically separated in actual implementation. In addition, the units or modules in the apparatus can be implemented in the form of processor calling software: for example, the apparatus includes a processor, the processor is connected with a memory, the memory stores instructions, and the processor calls the instructions stored in the memory to realize any of the above methods or realize the functions of each unit or module of the above apparatus, wherein the processor is a general processor such as a central processing unit (CPU) or a microprocessor, and the memory is a memory in the apparatus or a memory outside the apparatus. Alternatively, the units or modules in the apparatus can be implemented in the form of hardware circuit, and the functions of part or all of the units or modules can be realized by the design of hardware circuit. The above hardware circuit can be understood as one or more processors; for example, in one implementation, the above hardware circuit is an application-specific integrated circuit (ASIC), and the functions of part or all of the units or modules are realized by the design of the logical relationship of elements in the circuit; for another example, in another implementation, the above hardware circuit is a programmable logic device (PLD), and a field programmable gate array (FPGA) is taken as an example, which can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured by a configuration file, so as to realize the functions of part or all of the above units or modules. All units or modules of the above apparatus can be all implemented in the form of processor calling software, or all implemented in the form of hardware circuit, or part implemented in the form of processor calling software and the remaining part implemented in the form of hardware circuit.

[0349] In the embodiments of the present disclosure, the processor is a circuit with signal processing capability. In one implementation, the processor can be a circuit with instruction reading and running capability, such as a central processing unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a microprocessor), a digital signal processor (DSP), and the like. In another implementation, the processor can implement certain functions through a logical relationship of a hardware circuit, and the logical relationship of the hardware circuit is fixed or can be reconfigured. For example, the processor is a hardware circuit implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as an FPGA. In the reconfigurable hardware circuit, the processor loads a configuration document to implement the configuration of the hardware circuit. It can be understood that the processor loads instructions to implement the functions of the above part or all units or modules. In addition, the hardware circuit can also be designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), and the like.

[0350] FIG. 5A is a structural schematic diagram of a first device according to an embodiment of the present disclosure. As shown in FIG. 5A, the receiving end device 5100 can include a processing module 5101.

[0351] In some embodiments, the processing module 5101 is configured to perform channel measurement based on a first signal to determine channel measurement data, wherein the first signal is a signal obtained when a sensing signal is received; select one second AI model from a plurality of second artificial intelligence (AI) models, wherein each second AI model is used for device positioning or beam management in different environments; and determine the environment information based on the channel measurement data.

[0352] Optionally, the processing module 5101 is configured to perform at least one of other steps (for example, steps S2102, S2103, S2104, S2202, S2203a, and S2207, but not limited thereto) performed by the first device 5100 in any of the above methods. Details are not described herein again.

[0353] FIG. 5B is a structural schematic diagram of a first object according to an embodiment of the present disclosure. As shown in FIG. 5B, the second device 5200 can include a processing module 5201 and a transceiver module 5202.

[0354] In some embodiments, the processing module 5201 described above is configured to select one second artificial intelligence (AI) model that matches the environment information from a plurality of second AI models, wherein each of the second AI models is used for device positioning or beam management in different environments, and wherein the environment information is determined based on channel measurement data.

[0355] In some embodiments, the transceiver module 5202 described above is configured to send indication information to the first device, wherein the indication information is used to indicate an identity of the second AI model that matches the environment information.

[0356] Optionally, the processing module 5201 is configured to perform at least one of other steps (for example, steps S2104b and S2205, but not limited thereto) performed by the second device 5200 in any of the above methods, which will not be described here.

[0357] Optionally, the transceiver module 5202 is configured to perform at least one of the communication steps (for example, steps S2203b, S2204a, and S2206, but not limited thereto) performed by the second device 5200 in any of the above methods, which will not be described here.

[0358] In some embodiments, the sending module and / or the receiving module can be referred to as a transceiver module, and the sending module and the receiving module can be separate or integrated together. Optionally, the transceiver module can be mutually replaced with the transceiver.

[0359] In some embodiments, the processing module can be one module or can include a plurality of sub-modules. Optionally, the plurality of sub-modules perform all or part of the steps required to be performed by the processing module. Optionally, the processing module can be mutually replaced with the processor.

[0360] FIG. 6A is a structural schematic diagram of a communication device 6100 according to an embodiment of the present disclosure. The communication device 6100 can be a receiving end device (for example, a user equipment, a satellite, an Internet of Things device, etc.) or a first object (for example, a user equipment, an Internet of Things device, etc.), can be a chip, a chip system, or a processor supporting the receiving end device to implement any of the above methods, or can be a chip, a chip system, or a processor supporting the first object to implement any of the above methods. The communication device 6100 can be used to implement the methods described in the above method embodiments, and specific reference can be made to the descriptions in the above method embodiments.

[0361] As shown in FIG. 6A, the communication device 6100 includes one or more processors 6101. The processor 6101 can be a general processor or a special-purpose processor, etc., such as a baseband processor or a central processing unit. The baseband processor can be configured to process communication protocols and communication data, and the central processing unit can be configured to control a communication apparatus (e.g., a satellite, a terminal device, a terminal device chip, an environmental Internet of Things device, a TRP, etc.), execute programs, and process data of the programs. Optionally, the communication device 6100 is configured to perform any of the above methods. Optionally, the one or more processors 6101 are configured to invoke instructions to cause the communication device 6100 to perform any of the above methods.

[0362] In some embodiments, the communication device 6100 further includes one or more transceivers 6102. When the communication device 6100 includes one or more transceivers 6102, the transceiver 6102 performs at least one of the communication steps (e.g., steps S2101, steps S2201, steps S2203b, steps S2204a, steps S2206, but not limited to) in the above methods, and the processor 6101 performs at least one of the other steps (e.g., steps S2102, steps S2103, steps S2104, steps S2202, steps S2203a, steps S2204b, steps S2205, steps S2207, but not limited to) in the above methods. In optional embodiments, the transceiver can include a receiver and / or a transmitter, which can be separate or integrated together. Optionally, the terms transceiver, transceiving unit, transceiver, transceiving circuit, interface circuit, interface, etc. can be replaced with each other, and the terms transmitter, transmitting unit, transmitter, transmitting circuit, etc. can be replaced with each other, and the terms receiver, receiving unit, receiver, receiving circuit, etc. can be replaced with each other.

[0363] In some embodiments, the communication device 6100 further includes one or more memories 6103 for storing data. Optionally, all or part of the memory 6103 can also be outside the communication device 6100. In optional embodiments, the communication device 6100 can include one or more interface circuits 6104. Optionally, the interface circuit 6104 is connected to the memory 6103, and the interface circuit 6104 can be configured to receive data from the memory 6103 or other devices, and can be configured to send data to the memory 6103 or other devices. For example, the interface circuit 6104 can read data stored in the memory 6103 and send the data to the processor 6101.

[0364] The communication device 6100 described in the above embodiments can be a network device or a terminal, but the scope of the communication device 6100 described in the present disclosure is not limited thereto, and the structure of the communication device 6100 can not be limited by FIG. 6A. The communication device can be a standalone device or can be part of a larger device. For example, the communication device can be: 1) a standalone integrated circuit (IC), or a chip, or a chip system or subsystem; (2) a set of one or more ICs, which can optionally also include storage components for storing data, programs; (3) an ASIC, such as a modem; (4) a module that can be embedded in other devices; (5) a receiver, a terminal device, a smart terminal device, a cellular phone, a wireless device, a handset, a mobile unit, a vehicle-mounted device, a network device, a cloud device, an artificial intelligence device, and the like; (6) other devices, and the like.

[0365] FIG. 6B is a structural schematic diagram of a chip 6200 according to an embodiment of the present disclosure. For the case where the communication device 6100 is a chip or a chip system, the structural schematic diagram of the chip 6200 shown in FIG. 6B can be referred to, but is not limited thereto.

[0366] The chip 6200 includes one or more processors 6201. The chip 6200 is configured to perform any of the above methods.

[0367] In some embodiments, the chip 6200 further includes one or more interface circuits 6202. Optionally, the terms interface circuit, interface, transceiver pin, and the like can be replaced with each other. In some embodiments, the chip 6200 further includes one or more memories 6203 for storing data. Optionally, all or part of the memory 6203 can be outside the chip 6200. Optionally, the interface circuit 6202 is connected to the memory 6203, and the interface circuit 6202 can be configured to receive data from the memory 6203 or other devices, and the interface circuit 6202 can be configured to send data to the memory 6203 or other devices. For example, the interface circuit 6202 can read data stored in the memory 6203 and send the data to the processor 6201.

[0368] In some embodiments, the interface circuit 6202 performs at least one of the communication steps (for example, step S2101, step S2201, step S2203b, step S2204a, step S2206, but not limited to) of transmitting and / or receiving in the above method. The interface circuit 6202 performing the communication steps such as transmitting and / or receiving in the above method means that the interface circuit 6202 performs data interaction between the processor 6201, the chip 6200, the memory 6203, or the transceiver device. In some embodiments, the processor 6201 performs at least one of the other steps (for example, step S2102, step S2103, step S2104, step S2202, step S2203a, step S2204b, step S2205, step S2207, but not limited to).

[0369] The modules and / or devices described in each of the embodiments of the virtual device, the physical device, the chip, etc. can be combined or separated according to circumstances. Optionally, part or all of the steps can also be performed by a plurality of modules and / or devices in cooperation, which is not limited here.

[0370] The disclosure also proposes a storage medium, and the above storage medium stores instructions, which, when running on the communication device 5100, causes the communication device 5100 to perform any one of the above methods. Optionally, the above storage medium is an electronic storage medium. Optionally, the above storage medium is a computer readable storage medium, but is not limited to this, and it can also be a storage medium readable by other devices. Optionally, the above storage medium can be a non-transitory storage medium, but is not limited to this, and it can also be a transitory storage medium.

[0371] The disclosure also proposes a program product, which, when executed by the communication device 5100, causes the communication device 5100 to perform any one of the above methods. Optionally, the above program product is a computer program product.

[0372] The disclosure also proposes a computer program, which, when running on a computer, causes the computer to perform any one of the above methods.

[0373] Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure as disclosed herein. The disclosure is intended to cover any variations, uses or adaptive changes of the disclosure that follow the general principles of the disclosure and include known or customary practices in the art. The specification and examples are to be regarded as exemplary only, and the true scope and spirit of the disclosure are indicated by the following claims.

[0374] It should be understood that the present disclosure is not limited to the precise construction that has been described above and shown in the accompanying drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the present disclosure. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A model selection method characterized by, The method is performed by a first device, comprising: performing channel measurement based on a first signal to determine channel measurement data, wherein the first signal is a signal obtained when a sensing signal is received; selecting one second artificial intelligence (AI) model matching environment information from a plurality of second AI models, wherein each second AI model is used for device positioning or beam management in different environments, and wherein the environment information is determined based on the channel measurement data.

2. The method of claim 1, wherein, The method further comprises: inputting the channel measurement data into a first AI model to obtain the environment information output by the first AI model.

3. The method of claim 2, wherein, The method further comprises: collecting sample channel measurement data corresponding to different environment information; inputting the sample channel measurement data into an initial AI model to obtain estimated environment information output by the initial AI model; determining a loss function based on a difference between the estimated environment information and real environment information; training the initial AI model based on the loss function until the training is stopped when a stop condition is met, to obtain the first AI model.

4. The method of claim 3, wherein, The stop condition comprises at least one of the following: a number of training cycles is reached; the loss function is reduced to a fault tolerance range; the accuracy of an AI model after at least one round of training reaches a first value.

5. The method according to any one of claims 2-4, characterized in that, The method further comprises: sending the environment information to a second device; receiving indication information sent by the second device, wherein the indication information is used to indicate an identity of the second AI model matching the environment information.

6. The method of claim 1, wherein, The method further comprises: sending the channel measurement data to a second device; receiving indication information sent by the second device, wherein the indication information is used to indicate an identity of the second AI model matching the environment information.

7. The method according to claim 5 or 6, characterized in that, The selecting one second artificial intelligence (AI) model matching environment information from a plurality of second AI models comprises: selecting the second AI model indicated by the indication information from the plurality of second AI models.

8. The method according to any one of claims 2-4, characterized in that, The selecting one second artificial intelligence (AI) model matching environment information from a plurality of second AI models comprises: selecting one second AI model matching the environment information from the plurality of second AI models based on a correspondence between the second AI models and the environment information.

9. The method according to any one of claims 1 to 8, characterized in that, The environment information is used to indicate at least one of the following: an environment type; a scene type.

10. A model selection method characterized by comprising: The method is performed by a second device, comprising: selecting one second artificial intelligence (AI) model matching environment information from a plurality of second AI models, wherein each second AI model is used for device positioning or beam management in different environments, and wherein the environment information is determined based on channel measurement data; sending indication information to the first device, wherein the indication information is used to indicate an identity of the second AI model matching the environment information.

11. The method of claim 10, wherein, The method further comprises: receiving the channel measurement data sent by the first device; inputting the channel measurement data into a first AI model to obtain the environment information output by the first AI model.

12. The method of claim 11, wherein, The method further comprises: Collecting sample channel measurement data corresponding to different environmental information; Inputting the sample channel measurement data into an initial AI model to obtain estimated environmental information output by the initial AI model; Determining a loss function based on a difference between the estimated environmental information and real environmental information; Training the initial AI model based on the loss function until a stopping condition is met to stop training and obtain the first AI model.

13. The method of claim 12, wherein, The stopping condition includes at least one of the following: The number of training cycles is reached; The loss function is reduced to a fault tolerance range; The accuracy of the AI model after at least one round of training reaches a first value.

14. The method of claim 10, wherein, The method further includes: Receiving the environmental information sent by the first device.

15. The method according to any one of claims 10 to 14, characterized in that, The environmental information is used to indicate at least one of the following: environment type; scene type.

16. A first device, comprising: Comprise: a processing module configured to perform channel measurement based on a first signal to determine channel measurement data; wherein the first signal is a signal obtained when a received sensing signal is received; The processing module is further configured to select one of a plurality of second artificial intelligence (AI) models that matches the environmental information; wherein each of the second AI models is used for device positioning or beam management in different environments; wherein the environmental information is determined based on the channel measurement data.

17. A second device, comprising: Comprise: a processing module configured to select one of a plurality of second artificial intelligence (AI) models that matches the environmental information; wherein each of the second AI models is used for device positioning or beam management in different environments; wherein the environmental information is determined based on the channel measurement data; a transceiver module configured to send indication information to the first device; wherein the indication information is used to indicate the identity of the second AI model that matches the environmental information.

18. A communications device, characterized by The apparatus includes a processor and a memory, the memory having stored therein a computer program, and the processor executes the computer program stored in the memory to cause the apparatus to perform the method of any one of claims 1-9 or 10-15.

19. A communication system, characterized by Comprise: a third device configured to send a sensing signal; a first device configured to implement the method of any one of claims 1-9; a second device configured to implement the method of any one of claims 10-15.

20. A storage medium, the storage medium storing instructions, wherein, When the instructions are run on a communication device, the communication device is caused to perform the method of any one of claims 1-9 or 10-15.

21. A computer program product comprising a computer program, characterized in that, The computer program is executed by a processor to implement the method of any one of claims 1-9 or 10-15.

Images