Communication method and communication system

Abstract

The present disclosure relates to the technical field of communications, and provides a communication method and a communication system. The method comprises: a network device determines first information; and, on the basis of the first information, the network device determines an artificial intelligence (AI) model in a terminal, wherein the AI model comprises a first part, and the first part is used for adapting to different model input information and / or model output information. The AI model in the present disclosure is applicable to various types of model input information and / or model output information, and can match different input and output dimensions and types corresponding to different AI use cases or different resource configurations.

Description

Communication method and communication system TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of communication, and in particular to a communication method and a communication system. BACKGROUND

[0002] In recent years, artificial intelligence (AI) technology has made continuous breakthroughs in many fields. The continuous development of intelligent voice, computer vision and other fields not only brings a variety of applications to intelligent terminals, but also has wide application in education, transportation, home, medical care, retail, security and other fields, bringing convenience to people's life and promoting the industrial upgrading of various industries. AI technology is also accelerating the cross-penetration with other disciplines, and its development integrates knowledge from different disciplines, and also provides a new direction and method for the development of different disciplines.

[0003] SUMMARY

[0004] The present disclosure provides a communication method and a communication system. The present disclosure can determine an AI model in a terminal according to first information, the AI model is applicable to a plurality of model input information and / or model output information, can match different AI use cases or different input and output dimensions and types corresponding to resource configuration, can effectively reduce the number of AI model structures that need to be maintained on the terminal side, and reduce the complexity of the terminal.

[0005] The first aspect embodiment of the present disclosure provides a communication method, executed by a network device, the method comprising: determining first information; determining an AI model in a terminal according to the first information, the AI model comprising a first part, the first part being used to adapt to different model input information and / or model output information.

[0006] The second aspect embodiment of the present disclosure provides a communication method, executed by a terminal, the method comprising: determining first information; determining an AI model in the terminal according to the first information, the AI model comprising a first part, the first part being used to adapt to different model input information and / or model output information.

[0007] The third aspect embodiment of the present disclosure provides a communication method, comprising: a terminal sending first information to a network device; the network device receiving the first information sent by the terminal, and determining an AI model in the terminal according to the first information; wherein the AI model comprises a first part, and the first part is used to adapt to different model input information and / or model output information.

[0008] A fourth aspect of the present disclosure provides a network device, comprising: a processing module configured to determine first information; determine an AI model in a terminal according to the first information, the AI model comprising a first part, the first part being used to adapt to different model input information and / or model output information.

[0009] A fifth aspect of the present disclosure provides a terminal, comprising: a processing module configured to determine first information; determine an AI model in a terminal according to the first information, the AI model comprising a first part, the first part being used to adapt to different model input information and / or model output information.

[0010] A sixth aspect of the present disclosure provides a communication device, comprising: one or more processors; wherein the processor is configured to perform the method of the first aspect or the second aspect.

[0011] A seventh aspect of the present disclosure provides a communication system, comprising: a network device and a terminal, wherein the network device is configured to implement the method of the first aspect, and the terminal is configured to implement the method of any one of the second aspect.

[0012] An eighth aspect of the present disclosure provides a computer storage medium, wherein the computer storage medium stores computer executable instructions; the computer executable instructions are executed by a processor to implement the method of the first aspect or the second aspect.

[0013] A ninth aspect of the present disclosure provides a computer program product, comprising a computer program, the computer program being executed by a processor to implement the method of the first aspect or the second aspect.

[0014] Additional aspects and advantages of the present disclosure will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS

[0015] The above and / or additional aspects and advantages of the present disclosure will become apparent and be readily appreciated from the following description, including the accompanying drawings.

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

[0017] FIG. 2 is a schematic diagram of a flow of a communication method according to an embodiment of the present disclosure;

[0018] FIG. 3 is a schematic diagram of a flow of a communication method according to an embodiment of the present disclosure;

[0019] FIG. 4 is a flow diagram of a communication method according to an embodiment of the present disclosure;

[0020] FIG. 5 is a flow diagram of a communication method according to an embodiment of the present disclosure;

[0021] FIG. 6 is a flow diagram of a communication method according to an embodiment of the present disclosure;

[0022] FIG. 7 is a block diagram of a network device according to an embodiment of the present disclosure;

[0023] FIG. 8 is a block diagram of a terminal according to an embodiment of the present disclosure;

[0024] FIG. 9 is a structural diagram of a communication device according to an embodiment of the present disclosure;

[0025] FIG. 10 is a structural diagram of a chip according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0026] Embodiments of the present disclosure are described in detail below with reference to the accompanying drawings. Examples of the embodiments are shown in the drawings, and the same or similar reference numerals are used throughout to designate the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the drawings are exemplary and are intended to explain the present disclosure, and are not understood as limiting the present disclosure. It should be noted that the embodiments of the present disclosure and the features in the embodiments can be combined with each other without conflict.

[0027] To facilitate understanding, first introduce the terms related to the embodiments of the present disclosure.

[0028] Artificial Intelligence (AI) technology

[0029] AI technology has made continuous breakthroughs in many fields. The continuous development of intelligent voice, computer vision and other fields not only brings a variety of applications to intelligent terminals, but also has wide application in education, transportation, home, medical care, retail, security and many other fields, bringing convenience to people's lives and promoting the upgrading of various industries. AI technology is also accelerating the cross-penetration with other disciplines, and its development integrates knowledge from different disciplines, and also provides a new direction and method for the development of different disciplines.

[0030] In the 3GPP Release 18 stage, a research project on artificial intelligence technology in the wireless air interface is established in RAN1. The project aims to study how to introduce artificial intelligence technology in the wireless air interface, and to explore how artificial intelligence technology can assist in improving the transmission technology of the wireless air interface.

[0031] In the research of wireless AI, the application cases of AI technology include: AI-based CSI enhancement, AI-based beam management, AI-based positioning, and the like.

[0032] The embodiments of the present disclosure provide a communication method and a communication system.

[0033] In a first aspect, the embodiments of the present disclosure provide a communication method, performed by a network device, the method comprising: determining first information; determining an AI model in a terminal according to the first information, the AI model comprising a first part, the first part being used for adapting to different model input information and / or model output information.

[0034] The embodiments can determine the AI model in the terminal according to the first information, the AI model being applicable to a plurality of model input information and / or model output information, and being capable of matching different AI use cases or different input and output dimensions and types corresponding to resource configurations, thereby effectively reducing the number of AI model structures that need to be maintained at the terminal side and reducing the complexity of the terminal.

[0035] In combination with some embodiments of the first aspect, the first part comprises at least one of the following:

[0036] an input adaptation part, the input adaptation part being used for adapting to different model input information;

[0037] an output adaptation part, the output adaptation part being used for adapting to different model output information.

[0038] In combination with some embodiments of the first aspect, the input adaptation part comprises at least one first module, input parameter dimensions and / or input parameter types of the at least one first module being different, and output parameter dimensions and / or output parameter types of the at least one first module being the same.

[0039] In combination with some embodiments of the first aspect, the output adaptation part comprises at least one second module, input parameter dimensions and / or input parameter types of the at least one second module being the same, and output parameter dimensions and / or output parameter types of the at least one second module being different.

[0040] In combination with some embodiments of the first aspect, the AI model further comprises a second part, the second part being used for learning implicit features in a model training set and a relationship between the implicit features and a prediction target.

[0041] In combination with some embodiments of the first aspect, output parameters of the input adaptation part are used as input parameters of the second part.

[0042] In some embodiments of the first aspect, the output parameter dimension of the input adaptation part is the same as the input parameter dimension of the second part, and / or the output parameter type of the input adaptation part is the same as the input parameter type of the second part.

[0043] In some embodiments of the first aspect, the output parameter of the second part is as the input parameter of the output adaptation part.

[0044] In some embodiments of the first aspect, the input parameter dimension of the output adaptation part is the same as the output parameter dimension of the second part, and / or the input parameter type of the output adaptation part is the same as the output parameter type of the second part.

[0045] In some embodiments of the first aspect, the method further comprises: sending, to the terminal, model parameters corresponding to the AI model, wherein the model parameters comprise at least one of:

[0046] model parameters corresponding to the first part; and model parameters corresponding to the second part.

[0047] In some embodiments of the first aspect, the method further comprises: sending, to the terminal, second information; wherein the second information is used to indicate an input adaptation part and / or an output adaptation part corresponding to the model parameters.

[0048] In some embodiments of the first aspect, the second information comprises at least one of:

[0049] a first identifier used to determine an input adaptation part corresponding to the model parameters;

[0050] a second identifier used to determine an output adaptation part corresponding to the model parameters.

[0051] In some embodiments of the first aspect, the model input information comprises at least one of:

[0052] a model input dimension; and a model input type.

[0053] In some embodiments of the first aspect, the model output information comprises at least one of:

[0054] a model output dimension; and a model output type.

[0055] In some embodiments of the first aspect, the determining the first information comprises: receiving the first information sent by the terminal.

[0056] In some embodiments of the first aspect, the second network element is configured to provide computing power to a third network element and / or a terminal.

[0057] In some embodiments of the first aspect, the computing power is used for AI model training, and / or AI model inference, and / or perception service computing.

[0058] In some embodiments of the first aspect, the method further comprises determining a computing power policy of the second network element according to the computing power configuration and / or the computing power state.

[0059] In a second aspect, the embodiments of the present disclosure provide a communication method, executed by a terminal, the method comprising: determining first information; determining an AI model in the terminal according to the first information, the AI model comprising a first part, the first part being used for adapting to different model input information and / or model output information.

[0060] The embodiments can determine the AI model in the terminal according to the first information, the AI model being applicable to a plurality of model input information and / or model output information, and being able to match different AI use cases or different input and output dimensions and types corresponding to resource configurations, thereby effectively reducing the number of AI model structures that need to be maintained at the terminal side and reducing the complexity of the terminal.

[0061] In some embodiments of the second aspect, the first part comprises at least one of:

[0062] an input adaptation part, the input adaptation part being used for adapting to different model input information;

[0063] an output adaptation part, the output adaptation part being used for adapting to different model output information.

[0064] In some embodiments of the second aspect, the input adaptation part comprises at least one first module, input parameter dimensions and / or input parameter types of the at least one first module being different, and output parameter dimensions and / or output parameter types of the at least one first module being the same.

[0065] In some embodiments of the second aspect, the output adaptation part comprises at least one second module, input parameter dimensions and / or input parameter types of the at least one second module being the same, and output parameter dimensions and / or output parameter types of the at least one second module being different.

[0066] In some embodiments of the second aspect, the AI model further comprises a second part, the second part being used for learning implicit features in a model training set and a relationship between the implicit features and a prediction target.

[0067] In some embodiments of the second aspect, output parameters of the input adaptation part are used as input parameters of the second part.

[0068] In some embodiments of the second aspect, the output parameter dimension of the input adaptation part is the same as the input parameter dimension of the second part, and / or the output parameter type of the input adaptation part is the same as the input parameter type of the second part.

[0069] In some embodiments of the second aspect, the output parameter of the second part is used as the input parameter of the output adaptation part.

[0070] In some embodiments of the second aspect, the input parameter dimension of the output adaptation part is the same as the output parameter dimension of the second part, and / or the input parameter type of the output adaptation part is the same as the output parameter type of the second part.

[0071] In some embodiments of the second aspect, the method further comprises: receiving model parameters corresponding to the AI model sent by the network device; wherein the model parameters comprise at least one of:

[0072] the model parameters corresponding to the first part;

[0073] the model parameters corresponding to the second part.

[0074] In some embodiments of the second aspect, the method further comprises: receiving second information sent by the network device; and determining the input adaptation part and / or the output adaptation part corresponding to the model parameters according to the second information.

[0075] In some embodiments of the second aspect, the second information comprises at least one of:

[0076] a first identifier used to determine the input adaptation part corresponding to the model parameters;

[0077] a second identifier used to determine the output adaptation part corresponding to the model parameters.

[0078] In some embodiments of the second aspect, the model input information comprises at least one of:

[0079] a model input dimension; a model input type.

[0080] In some embodiments of the second aspect, the model output information comprises at least one of:

[0081] a model output dimension; a model output type.

[0082] In some embodiments of the second aspect, the method further comprises: sending the first information to the network device.

[0083] In a third aspect, the embodiments of the present disclosure provide a communication method, including: a terminal sending first information to a network device; the network device receiving the first information sent by the terminal, and determining an AI model in the terminal according to the first information; wherein the AI model includes a first part, and the first part is used to adapt to different model input information and / or model output information.

[0084] In a fourth aspect, the embodiments of the present disclosure provide a network device, including: a processing module configured to determine first information; determine an AI model in a terminal according to the first information, wherein the AI model includes a first part, and the first part is used to adapt to different model input information and / or model output information.

[0085] In a fifth aspect, the embodiments of the present disclosure provide a terminal, including: a processing module configured to determine first information; determine an AI model in a terminal according to the first information, wherein the AI model includes a first part, and the first part is used to adapt to different model input information and / or model output information.

[0086] In a sixth aspect, the embodiments of the present disclosure provide a communication device, including: one or more processors; wherein the processor is configured to execute the method according to the first aspect or the second aspect.

[0087] In a seventh aspect, the embodiments of the present disclosure provide a communication system, including: a network device and a terminal; the network device executes the method according to the first aspect, and the terminal executes the method according to the second aspect.

[0088] In an eighth aspect, the embodiments of the present disclosure provide a computer storage medium, wherein the computer storage medium stores computer executable instructions; the computer executable instructions are executed by a processor to implement the method according to the first aspect or the second aspect.

[0089] In a ninth aspect, the embodiments of the present disclosure provide a computer program product, including a computer program, wherein the computer program is executed by a processor to implement the method according to the first aspect or the second aspect.

[0090] In a tenth aspect, the embodiments of the present disclosure provide a chip or chip system. The chip or chip system includes a processing circuit configured to execute the method according to the first aspect or the second aspect.

[0091] It can be understood that the network device, the terminal, the communication system, and the storage medium are used to execute the method proposed in the embodiments of the present disclosure. Therefore, the beneficial effects achieved by the network device, the terminal, the communication system, and the storage medium can refer to the beneficial effects in the corresponding method, which will not be described here.

[0092] The embodiments of the present disclosure propose a communication method and a communication system. In some embodiments, the communication method and the information processing method, the information sending method, the information receiving method, and the like can be replaced with each other, the communication device and the information processing device, the information sending device, the information receiving device, and the like can be replaced with each other, and the information processing system, the communication system, the information sending system, the information receiving system, and the like can be replaced with each other.

[0093] The embodiments of the present disclosure are not exhaustive, but are only a part of the embodiments, and are not specific limitations on the protection scope of the present disclosure. In the case of no contradiction, each step in an embodiment can be implemented as an independent embodiment, and the steps can be combined arbitrarily, for example, the scheme after removing some steps in an embodiment can also be implemented as an independent embodiment, and the order of the steps in an embodiment can be exchanged arbitrarily, in addition, the optional implementation manners in an embodiment can be combined arbitrarily; in addition, the embodiments can be combined arbitrarily, for example, the steps of different embodiments or part or all of the steps of different embodiments can be combined arbitrarily, an embodiment can be combined with the optional implementation manners of other embodiments arbitrarily.

[0094] In the embodiments of the present disclosure, the terms and / or descriptions between the embodiments are consistent and can be referred to each other if there is no special description and logical conflict, and the technical features in different embodiments can be combined to form a new embodiment according to the logical relationship between them.

[0095] The terms used in the embodiments of the present disclosure are only for the purpose of describing the specific embodiments, and not as a limitation on the present disclosure.

[0096] In the embodiments of the present disclosure, unless otherwise specified, the elements expressed in singular form, such as “one”, “a”, “the”, “above”, “said”, “preceding”, “this”, and the like, can represent “one and only one”, or “one or more”, “at least one”, and the like. For example, in the case of using articles such as “a”, “an”, “the” in English, the noun after the article can be understood as singular expression, or as plural expression.

[0097] In the embodiments of the present disclosure, “a plurality of” means two or more.

[0098] In some embodiments, the terms “at least one of,” “at least one of,” “at least one of,” “one or more,” “a plurality of,” “multiple,” and the like can be replaced with each other.

[0099] In the description of the embodiments of the present disclosure, the description modes such as “at least one of A, B, and C”, “A and / or B and / or C”, and the like include any one of A, B, and C existing alone, and also include any combination of any number of A, B, and C, and each case can exist alone; for example, “at least one of A, B, and C” includes a case of A alone, a case of B alone, a case of C alone, a case of a combination of A and B, a case of a combination of A and C, a case of a combination of B and C, and a case of a combination of A and B and C; for example, A and / or B includes a case of A alone, a case of B alone, and a case of a combination of A and B.

[0100] In some embodiments, the description modes such as “A in a case, B in another case”, “in response to a case A, in response to another case B”, and the like can include the following technical solutions according to the cases: A is executed regardless of B, that is, A in some embodiments; B is executed regardless of A, that is, B in some embodiments; A and B are selectively executed, that is, A and B are selected from A and B to be executed in some embodiments; A and B are both executed, that is, A and B in some embodiments. When there are more branches of A, B, C, and the like, it is similar to the above.

[0101] The prefix words of "first", "second" and the like in the embodiments of the present disclosure are merely used to distinguish different description objects, and do not constitute 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 claims or embodiments, and should not constitute redundant limitation because of the use of the prefix words. For example, the description objects are "fields", and the ordinal words before "fields" in "first field" and "second field" do not limit the position or order between "fields". "First" and "second" do not limit whether the "fields" modified thereby are in the same message, nor do they limit the order of "first field" and "second field". For another example, the description objects are "levels", and the ordinal words before "levels" in "first level" and "second level" do not limit the priority between "levels". For another example, the quantity of the description objects is not limited by the ordinal words, and can be one or more. For example, "first device", wherein the quantity of "devices" can be one or more. In addition, the objects modified by different prefix words can be the same or different. For example, the description objects are "devices", 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 objects are "information", and "first information" and "second information" can be the same information or different information, and their contents can be the same or different.

[0102] In some embodiments, "including A", "containing A", "for indicating A", "carrying A" can be interpreted as directly carrying A, or indirectly indicating A.

[0103] In some embodiments, the terms of "in response to", "in response to determining", "in the case of", "when", "when", "if", "if" and the like can be replaced with each other.

[0104] In some embodiments, the terms of "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 of "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.

[0105] In some embodiments, an apparatus or the like can be interpreted as an entity, and can also be interpreted as virtual, and the name thereof is not limited to the name described in the embodiments, and the terms "apparatus", "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.

[0106] In some embodiments, a "network" can be interpreted as an apparatus (for example, an access network device, a core network device, and the like) included in the network.

[0107] In some embodiments, the terms "access network device (AN device)", "radio access network device (RAN device)", "base station (BS)", "radio base station", "fixed station", "node", "access point", "transmission point (TP)", "reception point (RP)", "transmission / reception point (TRP)", "panel", "antenna panel", "antenna array", "cell", "macro cell", "small cell", "femto cell", "pico cell", "sector", "cell group", "carrier", "component carrier", "bandwidth part (BWP)" and the like can be replaced with each other.

[0108] In some embodiments, the terms "terminal," "terminal device," "user equipment (UE)," "user terminal," "mobile station (MS)," "mobile terminal (MT)," "subscriber station," "mobile unit," "subscriber unit," "wireless unit," "remote unit," "mobile device," "wireless device," "wireless communication device," "remote device," "mobile subscriber station," "access terminal," "mobile terminal," "wireless terminal," "remote terminal," "handset," "user agent," "mobile client," "client," and so on can be replaced with each other.

[0109] In some embodiments, the access network device, the core network device, or the network device can be replaced with a terminal. For example, the embodiments of the present disclosure can also be applied to a structure in which communication between the access network device, the core network device, or the network device and the terminal is replaced with communication between a plurality of terminals (for example, also referred to as device-to-device (D2D), vehicle-to-everything (V2X), and so on). In this case, the terminal can also be configured to have all or part of the functions of the access network device. In addition, the language of "uplink," "downlink," and so on can also be replaced with language corresponding to the inter-terminal communication (for example, "side"). For example, the uplink channel, the downlink channel, and so on can be replaced with the side channel, and the uplink, the downlink, and so on can be replaced with the side link.

[0110] In some embodiments, the terminal can be replaced with the access network device, the core network device, or the network device. In this case, the access network device, the core network device, or the network device can also be configured to have all or part of the functions of the terminal.

[0111] In some embodiments, the data, information, etc. can be obtained in compliance with the laws and regulations of the country where the location is situated.

[0112] In some embodiments, the data, information, etc. can be obtained after obtaining the consent of the user.

[0113] In some embodiments, the threshold mentioned in the embodiments can be a numerical value, a constant, or some fixed value, etc.

[0114] In some embodiments, the positioning and the location mentioned in the embodiments can have the same meaning.

[0115] In addition, each element, each row, or each column in the table of the embodiments of the disclosure can be implemented as an independent embodiment, and any element, any row, or any column combination can also be implemented as an independent embodiment.

[0116] The correspondence shown in each table in the disclosure can be configured or predefined. The values of the information in each table are only examples, and other values can be configured, and the disclosure is not limited. When configuring the correspondence between the information and each parameter, it is not necessarily required to configure all the correspondences shown in each table. For example, the correspondence shown in some rows in the table in the disclosure can not be configured. For another example, the above table can be appropriately deformed, adjusted, etc., such as splitting, merging, etc. The name of the parameter shown in the title of each table in the above can also use other names understandable by the communication device, and the value or representation of the parameter can also use other values or representations understandable by the communication device. Each table in the above can also use other data structures when implemented, such as array, queue, container, stack, linear table, pointer, linked list, tree, graph, structure, class, heap, hash table, etc.

[0117] The predefinition in the disclosure can be understood as definition, predefinition, storage, pre-storage, pre-negotiation, pre-configuration, solidification, or pre-burning.

[0118] The communication method and the communication system provided by the disclosure will be described in detail below with reference to the accompanying drawings.

[0119] FIG. 1 shows a structure diagram of a communication system according to an embodiment of the disclosure, as shown in FIG. 1, the system architecture can include a network device 11 and a terminal 12.

[0120] In some embodiments, the network device 11 can be an entity for transmitting or receiving a signal. For example, it can include a core network element, a passive Internet of Things server, a communication satellite, an evolved NodeB (eNB), a transmission reception point (TRP), a next generation NodeB (gNB) in an NR system, a base station in other future mobile communication systems, or an access node in a wireless fidelity (WiFi) system, a terminal, a passive Internet of Things device, and the like. Embodiments of the present disclosure do not limit the specific technology and specific device form adopted by the network device 11. The network device 11 provided by the embodiments of the present disclosure can be composed of a central unit (CU) and a distributed unit (DU), wherein the CU can also be referred to as a control unit. The CU-DU structure can split the protocol layer of the network device, for example, the base station, and place part of the protocol layer functions in the CU for centralized control, and the remaining part or all of the protocol layer functions are distributed in the DU and controlled by the CU.

[0121] In some embodiments, the terminal 12 can be referred to as a terminal device, a user equipment, a mobile station (MS), a mobile terminal (MT), an NB-IoT terminal, and the like. The terminal 12 can also be a communication-capable automobile, a smart automobile, a mobile phone, a wearable device, a tablet computer (Pad), a computer with wireless transceiver function, a virtual reality device, an augmented reality device, a wireless terminal device in industrial control, a wireless terminal device in self-driving, a wireless terminal device in remote medical surgery, a wireless terminal device in smart grid, a wireless terminal device in transportation safety, a wireless terminal device in smart city, a wireless terminal device in smart home, and the like. Embodiments of the present disclosure do not limit the specific technology and specific device form adopted by the terminal 12.

[0122] It can be understood that the communication system described in the embodiments of the present disclosure is for more clearly illustrating the technical solutions of the embodiments of the present disclosure, and does not constitute a limitation on the technical solutions proposed by the embodiments of the present disclosure. Those skilled in the art can know that, with the evolution of system architecture and the appearance of new business scenarios, the technical solutions proposed by the embodiments of the present disclosure are also applicable to similar technical problems.

[0123] The following embodiments of the present disclosure can be applied to the communication system shown in FIG. 1 or part of the subject, but are not limited thereto. The subjects shown in FIG. 1 are illustrative, and the communication system can include all or part of the subjects in FIG. 1, or other subjects other than FIG. 1. The number and form of each subject is arbitrary, and the connection relationship between the subjects is illustrative. The subjects can be connected or not connected, and the connection can be in any way, can be direct connection or indirect connection, can be wired connection or wireless connection.

[0124] Embodiments of the present disclosure can be applied to satellite communication, 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 NR, 6th generation mobile communication system (6G), Future Radio Access (FRA), New-Radio Access Technology (RAT), New Radio (NR), New radio access (NX), Future generation radio access (FX), Global System for Mobile communications (GSM (registered trademark)), CDMA2000, Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi (registered trademark)), IEEE 802.16 (WiMAX (registered trademark)), IEEE 802.20, Ultra-WideBand (UWB), Bluetooth (Bluetooth (registered trademark)), Public Land Mobile Network (PLMN) network, Device-to-Device (D2D) system, Machine to Machine (M2M) system, Internet of Things (IOT) system, Vehicle-to-Everything (V2X), system using other communication methods, next-generation system expanded based thereon, and the like. Further, a plurality of systems can be applied in combination (for example, combination of LTE or LTE-A and 5G, and the like).

[0125] As for the deployment of the AI model, one way is that the network side trains the AI model and then transmits the AI model to the terminal. However, in this case, the terminal side needs to compile the AI model, and the hardware requirement of the terminal is relatively high. Further, the AI model can be divided into two parts, one part is the structure of the AI model, and the other part is the model parameters corresponding to the model structure. In order to reduce the complexity of the terminal deploying the AI model, the terminal side can deploy the model structure in advance. The network side can transmit the corresponding model parameters to the model structure deployed by the terminal. The terminal applies the model parameters of the network side to the corresponding model structure to form a complete AI model.

[0126] However, due to different resource configurations or different use cases in actual use cases, the requirements of the parameter dimensions and types of the input and output of the AI model are different. If corresponding AI model structures are designed for different parameter dimensions and types of the input and output, the terminal needs to maintain more AI model structures, which increases the complexity of the terminal.

[0127] To solve the above problems, the embodiment can determine the AI model in the terminal according to the first information, which is suitable for multiple model input information and / or model output information, can match different input and output dimensions and types corresponding to different AI use cases or resource configurations, and can effectively reduce the number of AI model structures that need to be maintained by the terminal, thereby reducing the complexity of the terminal.

[0128] Further, in order to illustrate the specific execution process of the above communication system, FIG. 2 shows a schematic diagram of a communication method according to an embodiment of the present disclosure. The method is applied to the above communication system, as shown in FIG. 2, and can include the following steps:

[0129] Step S201, the terminal sends first information to the network device.

[0130] In some embodiments, the network device receives the first information sent by the terminal.

[0131] In some embodiments, the first information can be used to determine the AI model in the terminal.

[0132] In some embodiments, the first information can be carried by at least one of the following:

[0133] Radio Resource Control (RRC) message, Medium Access Control (MAC) control element (Control Element, CE), uplink control information (Uplink Control Information, UCI), etc.

[0134] In some embodiments, after determining the model structure of the AI model, the terminal can send first information to the network device to indicate the model structure of the AI model in the terminal, so that the network device can send the model parameters corresponding to the model structure.

[0135] In some embodiments, the terminal can actively send the first information to the network device, or send the first information to the network device according to the indication of the network device, etc.

[0136] In some embodiments, the terminal can also determine the first information with the network device through offline negotiation, such as determining the first information based on private predefinition, and then determining the model structure of the AI model in the terminal through the first information.

[0137] In some embodiments, the terminal can also determine the first information with the network device through standard predefinition, such as determining the first information based on protocol predefinition, and then determining the model structure of the AI model in the terminal through the first information.

[0138] In step S202, the network device determines the AI model in the terminal according to the first information.

[0139] In some embodiments, the AI model includes a first part, and the first part is used to adapt to different model input information and / or model output information. In some examples, the model input information can include model input dimensions and / or model input types, etc. In some examples, the model output information can include model output dimensions and / or model output types, etc.

[0140] The AI model determined in this embodiment, wherein the first part can flexibly adapt to different model input information and / or model output information, so that the AI model has high generalization ability and flexibility, improves the wide applicability of the AI model, effectively utilizes different forms of input data, and generates output information that meets specific needs. The AI model is suitable for various model input information and / or model output information, can match different AI use cases or different input and output dimensions and types corresponding to resource configuration, can effectively reduce the number of AI model structures that need to be maintained on the terminal side, and reduces the complexity of the terminal.

[0141] In some embodiments, the first part can include at least one of A1 to B1:

[0142] A1, an input adaptation part, the input adaptation part is used to adapt to different model input information. In some examples, the input adaptation part can be an adaptation part for model input, which can adapt to different model input dimensions and / or model input types, etc.

[0143] B1, an output adaptation part, the output adaptation part being configured to adapt different model output information. In some examples, the output adaptation part can be an adaptation part for model output, which can adapt different model output dimensions and / or model output types, etc.

[0144] In some embodiments, the input adaptation part can include at least one first module, wherein the input parameter dimensions and / or input parameter types of the at least one first module are different, and the output parameter dimensions and / or output parameter types of the at least one first module are the same.

[0145] In some examples, different input parameter dimensions and / or input parameter types can correspond to different first modules.

[0146] For example, the input adaptation part can include three first modules, namely first module 1, first module 2 and first module 3, the input parameter type corresponding to the first module 1 can be type 4, the input parameter type corresponding to the first module 2 can be type 3, and the input parameter type corresponding to the first module 3 can be type 5. After data processing inside these first modules, the output parameter types corresponding to the first module 1, the first module 2 and the first module 3 can all be type 2.

[0147] In some embodiments, the output adaptation part can include at least one second module, wherein the input parameter dimensions and / or input parameter types of the at least one second module are the same, and the output parameter dimensions and / or output parameter types of the at least one second module are different.

[0148] In some examples, different output parameter dimensions and / or output parameter types can correspond to different second modules.

[0149] For example, the output adaptation part can include three second modules, namely second module 1, second module 2 and second module 3, the input parameter types corresponding to the second module 1, the second module 2 and the second module 3 can all be type 2, and after data processing inside these second modules, the output parameter type corresponding to the second module 1 can be type 4, the input parameter type corresponding to the second module 2 can be type 3, and the input parameter type corresponding to the second module 3 can be type 5.

[0150] In some embodiments, the AI model can further include a second part, which is configured to learn the implicit features in the model training set and the relationship between the implicit features and the prediction target. In some examples, the second part can be referred to as a core part, a main part, a central part, a key part, etc., and the present embodiment is not limited thereto.

[0151] In some examples, the second part is necessary for the AI model. As shown in FIG. 3, the second part can be a core part, where the input parameter dimension and the output parameter dimension of the core part are fixed, for example, the input parameter dimension of the core part can be a vector of [X*Y*Z], and the output parameter dimension can be a vector of [M*N]. The structure of the core part of different models can be the same, but when the data set for model training is different, the model parameters corresponding to the core part structure are different.

[0152] In some embodiments, the model structure is determined by combining the first part and the second part, which can reduce the number of model structures that need to be maintained, and reduce the complexity of the terminal.

[0153] In some embodiments, the output parameter of the input adaptation part of A1 is used as the input parameter of the second part. For example, the output parameter of the input adaptation part is used as the input parameter of the second part, and data processing is performed in the second part.

[0154] In some embodiments, the output parameter dimension of the input adaptation part of A1 is the same as the input parameter dimension of the second part, and / or the output parameter type of the input adaptation part is the same as the input parameter type of the second part.

[0155] In some embodiments, the output parameter of the second part is used as the input parameter of the output adaptation part of B1. For example, the output parameter of the second part is used as the input parameter of the output adaptation part, and data processing is performed in the output adaptation part.

[0156] In some embodiments, the input parameter dimension of the output adaptation part of B1 is the same as the output parameter dimension of the second part, and / or the input parameter type of the output adaptation part is the same as the output parameter type of the second part.

[0157] In some embodiments, the parameters of the input adaptation part and the output adaptation part correspond to the input and output parameters of the second part, which can avoid errors caused by data mismatch, ensure that the model can accurately understand and process input information, and thus improve the accuracy of prediction or classification.

[0158] In some examples, as shown in FIG. 3, the output parameter dimension of the input adaptation part and the output parameter dimension of the second part can both be a vector of [X*Y*Z]; the input parameter dimension of the output adaptation part and the output parameter dimension of the second part can both be a vector of [M*N].

[0159] In some embodiments, the network device sends the model parameters corresponding to the AI model to the terminal. Correspondingly, in some embodiments, the terminal receives the model parameters corresponding to the AI model sent by the network device.

[0160] In some embodiments, the model parameters can include at least one of the following A2 to B2:

[0161] A2, the model parameters corresponding to the first part, such as the model parameters corresponding to the input adaptation part and / or the output adaptation part.

[0162] B2, the model parameters corresponding to the second part.

[0163] In some embodiments, the network device sends the model parameters to the terminal, and can only send part of the parameters, for example, can only send the parameters of the input adaptation part, the core part or the output adaptation part. For example, the terminal has currently received a model parameter package containing the model parameters corresponding to the first part and the model parameters corresponding to the second part, and when the model parameters need to be updated, the model parameters of the core part can be kept unchanged, and only the parameters of the input adaptation part and / or the output adaptation part are updated.

[0164] In some embodiments, the network device sends second information to the terminal; wherein the second information is used to indicate the input adaptation part and / or the output adaptation part corresponding to the model parameters sent by the network device. Correspondingly, in some embodiments, the terminal receives the second information sent by the network device.

[0165] In some embodiments, the second information can include at least one of the following A3 to B3:

[0166] A3, a first identifier, the first identifier is used to determine the input adaptation part corresponding to the model parameters.

[0167] B3, a second identifier, the second identifier is used to determine the output adaptation part corresponding to the model parameters.

[0168] In some examples, the terminal determines the input adaptation part and / or the output adaptation part corresponding to the model parameters sent by the network device according to the second information.

[0169] In some embodiments, when the terminal deploys multiple input adaptation parts or multiple output adaptation parts, the network device needs to indicate to the terminal which input adaptation part or which output adaptation part the model parameters correspond to when the network sends the corresponding model parameters to the terminal. For example, as shown in FIG. 3, if the first identifier points to the #2 input adaptation part and the second identifier points to the #3 output adaptation part, the second information indicates that the model parameters correspond to the model composed of the #2 input adaptation part + the core part + the #3 output adaptation part.

[0170] In some embodiments, the model input dimension and / or the model output dimension determine the size or complexity of the data structure in space expected to be received by the model, and the model input type and / or the model output type can indicate the specific form of the data or the nature of the data, so as to ensure that the data is correctly interpreted and processed.

[0171] The AI model can be used in various application scenarios, such as terminal positioning based on the AI model, or beam management based on the AI model, and the like. In some embodiments, taking terminal positioning based on the AI model as an example, the AI model deployed in the terminal can analyze and determine the positioning information of the terminal by utilizing various sensors and communication signals.

[0172] For example, first, the AI model, such as the model structure of the AI model, including the first part and the second part, can be determined at the network device side by the terminal reporting to the network device, which can be specifically referred to the above embodiments. After the network device determines the AI model, the corresponding model parameters (i.e., the parameters obtained by training the AI model) can be sent to the terminal, so that the terminal is deployed with the trained AI model. Then, various sensors and communication modules are used to collect wireless signal data, built-in sensor data, environmental perception data, multi-modal data, and the like. The data is input into the trained AI model to obtain the positioning information of the terminal.

[0173] In some examples, the wireless signal data includes, but is not limited to, signal strength (RSSI), time of arrival (TOA), time difference of arrival (TDOA), angle of arrival (AoA), and the like, which can be received from multiple base stations; the built-in sensor data can include accelerometer, gyroscope, magnetometer, barometer, and the like, which can provide device movement and attitude information.

[0174] In some examples, target features can be extracted from the above data for visual positioning, wherein the target features can include signal features, sensor features, environmental features, and the like. In some examples, the signal features can include signal strength variation, frequency response, Doppler shift, and the like; the sensor features include acceleration, rotation, movement direction and speed, and the like; the environmental features include obstacle influence, reflection and diffraction patterns, and the like.

[0175] In some examples, during the training of the AI model, the network device can use a large number of labeled data sets to train the target AI model to learn to map different sensor inputs to accurate position information. The data set should contain the true coordinates of the terminal position in different environments and the corresponding signal and sensor data, for example, using a deep neural network, a convolutional neural network, or a recurrent neural network to predict the position coordinates.

[0176] In some examples, the terminal submits target features from the collected data and inputs the extracted target features into the AI model, and the model outputs the predicted position of the terminal.

[0177] In some examples, the positioning result output by the AI model can be fed back to the AI model for continuous learning and optimization to improve the stability of positioning.

[0178] The communication method related to the embodiment can include at least one of steps S201-S202. For example, step S201 can be implemented as an independent embodiment, and step S202 can be implemented as an independent embodiment. In addition, part or all of the steps S201-S202 can be combined to implement an independent embodiment, which is not limited in this embodiment.

[0179] The embodiment can determine the AI model in the terminal according to the first information. The AI model is applicable to a plurality of model input information and / or model output information, can match different AI use cases or different input and output dimensions and types corresponding to resource configuration, can effectively reduce the number of AI model structures that need to be maintained on the terminal side, and reduces the complexity of the terminal.

[0180] In order to illustrate the specific execution process of the network device, FIG. 4 shows a flowchart of a communication method according to an embodiment of the present disclosure. It can include the following steps, which are executed by the network device.

[0181] Step S301, the network device determines the first information.

[0182] In some embodiments, the first information can be used to determine the AI model in the terminal.

[0183] In some embodiments, the network device receives the first information sent by the terminal.

[0184] In some embodiments, the terminal can also determine the first information with the network device through offline negotiation, such as determining the first information by the terminal and the network device based on a private predefinition, and then determining the AI model in the terminal through the first information.

[0185] In some embodiments, the terminal can also determine the first information with the network device through a standard predefinition, such as determining the first information by the terminal and the network device based on a protocol predefinition, and then determining the AI model in the terminal through the first information.

[0186] Step S302, the network device determines the AI model in the terminal according to the first information.

[0187] In some embodiments, the AI model includes a first part, and the first part is used to adapt to different model input information and / or model output information.

[0188] In some embodiments, the first part includes at least one of the following:

[0189] The input adaptation part is used to adapt to different model input information;

[0190] The output adaptation part is configured to adapt different model output information.

[0191] In some embodiments, the input adaptation part includes at least one first module, the input parameter dimension and / or the input parameter type of the at least one first module are different, and the output parameter dimension and / or the output parameter type of the at least one first module are the same.

[0192] In some embodiments, the output adaptation part includes at least one second module, the input parameter dimension and / or the input parameter type of the at least one second module are the same, and the output parameter dimension and / or the output parameter type of the at least one second module are different.

[0193] In some embodiments, the AI model further includes a second part, the second part is configured to learn the implicit features in the model training set and the relationship between the implicit features and the prediction target.

[0194] In some embodiments, the output parameter of the input adaptation part is used as the input parameter of the second part.

[0195] In some embodiments, the output parameter dimension of the input adaptation part is the same as the input parameter dimension of the second part, and / or the output parameter type of the input adaptation part is the same as the input parameter type of the second part.

[0196] In some embodiments, the output parameter of the second part is used as the input parameter of the output adaptation part.

[0197] In some embodiments, the input parameter dimension of the output adaptation part is the same as the output parameter dimension of the second part, and / or the input parameter type of the output adaptation part is the same as the output parameter type of the second part.

[0198] In some embodiments, the model input information includes at least one of the following:

[0199] The model input dimension; the model input type.

[0200] In some embodiments, the model output information includes at least one of the following:

[0201] The model output dimension; the model output type.

[0202] In some embodiments, the network device sends the model parameters corresponding to the AI model to the terminal.

[0203] The model parameters include at least one of the following:

[0204] The model parameters corresponding to the first part; the model parameters corresponding to the second part.

[0205] In some embodiments, the network device sends second information to the terminal; wherein the second information is used to indicate the input adaptation part and / or the output adaptation part corresponding to the model parameter.

[0206] In some embodiments, the second information includes at least one of the following:

[0207] The first identifier is used to determine the input adaptation part corresponding to the model parameter;

[0208] The second identifier is used to determine the output adaptation part corresponding to the model parameter.

[0209] For a detailed description of the specific examples in this embodiment, please refer to the corresponding description of the embodiments in FIGS. 1-4, which will not be repeated here.

[0210] The communication method related to the present embodiment can include at least one of steps S301-S302. For example, step S301 can be implemented as an independent embodiment, and step S302 can be implemented as an independent embodiment. In addition, part or all of the steps S301-S302 can be combined to implement an independent embodiment, which is not limited in the present embodiment.

[0211] In the present embodiment, the network device determines the AI model in the terminal according to the first information. The AI model is applicable to multiple model input information and / or model output information, and can match different AI use cases or different input and output dimensions and types corresponding to different resource configurations.

[0212] FIG. 5 shows a flow diagram of a communication method according to an embodiment of the present disclosure. As shown in FIG. 5, the method is performed by a terminal and can include the following steps.

[0213] Step S401, the terminal determines the first information.

[0214] In some embodiments, the first information can be used to determine the AI model in the terminal.

[0215] In some embodiments, the terminal can also determine the first information with the network device through offline negotiation, such as determining the first information by the terminal and the network device based on private predefinition, and then determining the AI model in the terminal through the first information.

[0216] In some embodiments, the terminal can also determine the first information with the network device through standard predefinition, such as determining the first information by the terminal and the network device based on protocol predefinition, and then determining the AI model in the terminal through the first information.

[0217] In some embodiments, the terminal sends the first information to the network device.

[0218] Step S402, the terminal determines the AI model in the terminal according to the first information.

[0219] In some embodiments, the AI model comprises a first part, and the first part is used to adapt different model input information and / or model output information.

[0220] In some embodiments, the first part comprises at least one of the following:

[0221] an input adaptation part, the input adaptation part being used to adapt different model input information;

[0222] an output adaptation part, the output adaptation part being used to adapt different model output information.

[0223] In some embodiments, the input adaptation part comprises at least one first module, the input parameter dimension and / or the input parameter type of the at least one first module being different, and the output parameter dimension and / or the output parameter type of the at least one first module being the same.

[0224] In some embodiments, the output adaptation part comprises at least one second module, the input parameter dimension and / or the input parameter type of the at least one second module being the same, and the output parameter dimension and / or the output parameter type of the at least one second module being different.

[0225] In some embodiments, the AI model further comprises a second part, and the second part is used to learn the implicit features in the model training set and the relationship between the implicit features and the prediction target.

[0226] In some embodiments, the output parameter of the input adaptation part is used as the input parameter of the second part.

[0227] In some embodiments, the output parameter dimension of the input adaptation part is the same as the input parameter dimension of the second part, and / or the output parameter type of the input adaptation part is the same as the input parameter type of the second part.

[0228] In some embodiments, the output parameter of the second part is used as the input parameter of the output adaptation part.

[0229] In some embodiments, the input parameter dimension of the output adaptation part is the same as the output parameter dimension of the second part, and / or the input parameter type of the output adaptation part is the same as the output parameter type of the second part.

[0230] In some embodiments, the model input information comprises at least one of the following:

[0231] a model input dimension; a model input type.

[0232] In some embodiments, the model output information comprises at least one of the following:

[0233] Model output dimension; model output type.

[0234] In some embodiments, the terminal receives the model parameters corresponding to the structural model sent by the network device.

[0235] The model parameters include at least one of the following:

[0236] The model parameters corresponding to the first part; the model parameters corresponding to the second part.

[0237] In some embodiments, the terminal receives the second information sent by the network device; and the terminal determines the input adaptive part and / or the output adaptive part corresponding to the model parameters according to the second information.

[0238] In some embodiments, the second information includes at least one of the following:

[0239] The first identifier is used to determine the input adaptive part corresponding to the model parameters.

[0240] The second identifier is used to determine the output adaptive part corresponding to the model parameters.

[0241] For a detailed description of the specific examples in this embodiment, please refer to the corresponding description of the embodiments in FIGS. 1-5, which will not be repeated here.

[0242] The communication method related by the embodiment can include at least one of steps S401-S402. For example, step S401 can be implemented as an independent embodiment, and step S402 can be implemented as an independent embodiment. In addition, part or all of the steps of steps S401-S402 can be combined to implement an independent embodiment, and the embodiment is not limited in this regard.

[0243] The communication method provided by the embodiment can determine the AI model in the terminal according to the first information, which is suitable for multiple model input information and / or model output information, and can match different AI use cases or different input and output dimensions and types corresponding to different resource configurations.

[0244] FIG. 6 is an interaction diagram of a communication method according to an embodiment of the present disclosure, as shown in FIG. 6, the embodiment of the present disclosure relates to a communication method, which includes:

[0245] Step S501, the network device determines the AI model in the terminal according to the first information.

[0246] In some embodiments, the terminal determines the AI model in the terminal according to the first information.

[0247] In some embodiments, the AI model comprises a first part, which is used to adapt different model input information and / or model output information.

[0248] In some embodiments, the network device receives the first information sent by the terminal.

[0249] In some embodiments, the terminal sends the first information to the network device.

[0250] In some embodiments, the terminal and the network device agree on the understanding of the input adaptation part, the core part and the output adaptation part of the model. Specifically, the terminal and the network device reach the purpose of understanding by offline negotiation or the terminal reports to the network or by standard predefined way. The offline negotiation is a private negotiation between the terminal and the network device, which does not go through the signaling process, i.e., based on private predefined way.

[0251] In some embodiments, the first part comprises at least one of:

[0252] an input adaptation part, which is used to adapt different model input information;

[0253] an output adaptation part, which is used to adapt different model output information.

[0254] In some embodiments, the input adaptation part comprises at least one first module, the input parameter dimension and / or the input parameter type of the at least one first module are different, and the output parameter dimension and / or the output parameter type of the at least one first module are the same.

[0255] In some embodiments, the input adaptation part comprises one or more model structures, the input parameter dimension and / or the input parameter type of the one or more model structures are different, the output parameter dimension and / or the output parameter type of the one or more model structures are the same, and the model output parameter dimension and / or the output parameter type of the input adaptation part are the same as the input parameter dimension and / or the input parameter type of the core part.

[0256] In some embodiments, the output adaptation part comprises at least one second module, the input parameter dimension and / or the input parameter type of the at least one second module are the same, and the output parameter dimension and / or the output parameter type of the at least one second module are different.

[0257] In some embodiments, the output adaptation part comprises one or more model structures, the output parameter dimension and / or the output parameter type of the one or more model structures are different, but the input parameter dimension and / or the input parameter type are the same, and the model input parameter dimension and / or the input parameter type of the output adaptation part are the same as the output parameter dimension and / or the output parameter type of the core part.

[0258] In some embodiments, the AI model further comprises a second part, which is used to learn the hidden features in the model training set and the relationship between the hidden features and the prediction target.

[0259] In some embodiments, the model comprises at least two parts, including an input adaptation part, a core part, and an output adaptation part, and the model must contain the core part.

[0260] In some embodiments, the structure of the core part of different models can be the same. When the training data set is different, the model parameters corresponding to the core part structure are different. The input parameter dimension and the output parameter dimension of the core part are fixed, for example, the input parameter dimension of the core part can be a [X*Y*Z] vector, and the output parameter dimension can be an [M*N] vector.

[0261] In some embodiments, the output parameters of the input adaptation part are used as the input parameters of the second part.

[0262] In some embodiments, the output parameter dimension of the input adaptation part is the same as the input parameter dimension of the second part, and / or the output parameter type of the input adaptation part is the same as the input parameter type of the second part.

[0263] In some embodiments, the output parameters of the second part are used as the input parameters of the output adaptation part.

[0264] In some embodiments, the input parameter dimension of the output adaptation part is the same as the output parameter dimension of the second part, and / or the input parameter type of the output adaptation part is the same as the output parameter type of the second part.

[0265] In some embodiments, the model input information includes at least one of the following:

[0266] The model input dimension; the model input type.

[0267] In some embodiments, the model output information includes at least one of the following:

[0268] The model output dimension; the model output type.

[0269] In some embodiments, the input adaptation part and the output adaptation part include one or more model structures, so as to adapt to different input parameter dimensions and / or input parameter types, or output parameter dimensions and / or output parameter types.

[0270] In some embodiments, different input parameter dimensions and / or input parameter types and output parameter dimensions and / or output parameter types can correspond to different model structures.

[0271] In step S502, the network device sends the model parameters corresponding to the AI model to the terminal.

[0272] In some embodiments, the terminal receives the model parameters corresponding to the AI model sent by the network device.

[0273] The model parameters include at least one of the following:

[0274] The model parameters corresponding to the first part; and the model parameters corresponding to the second part.

[0275] In some embodiments, the network device sends the model parameters to the terminal, and only part of the parameters can be sent, for example, only the parameters of the input adaptation part, the core part, or the output adaptation part can be sent. In a complete model parameter package, the core part is necessary. For the case of updating the model parameters, if the terminal has received a model parameter package containing the parameters corresponding to the core part, the input adaptation part, and the output adaptation part, the core part can remain unchanged during the update, and only the input adaptation part and the output adaptation part can be updated.

[0276] In some embodiments, the network device sends the second information to the terminal.

[0277] The second information is used to indicate the input adaptation part and / or the output adaptation part corresponding to the model parameters.

[0278] In some embodiments, the terminal receives the second information sent by the network device; and the terminal determines the input adaptation part and / or the output adaptation part corresponding to the model parameters according to the second information.

[0279] In some embodiments, the second information includes at least one of the following:

[0280] The first identifier is used to determine the input adaptation part corresponding to the model parameters.

[0281] The second identifier is used to determine the output adaptation part corresponding to the model parameters.

[0282] In some embodiments, the model parameters need to correspond to the model structure, so it is necessary to indicate which specific model structure the model parameters correspond to, including the input adaptation part and / or the output adaptation part.

[0283] In some embodiments, when the terminal deploys multiple input adaptation parts or multiple output adaptation parts, the network device needs to indicate to the terminal which input adaptation part or which output adaptation part the model parameters correspond to when sending the corresponding model parameters to the terminal. For example, it is indicated that the model parameters can correspond to a model composed of #2 input adaptation part + core part + #3 input adaptation part.

[0284] In some embodiments, a model structure supporting multiple input parameter dimensions and / or input parameter types and output parameter dimensions and / or output parameter types is proposed to match different input and output dimensions corresponding to different AI use cases or different configurations, thereby reducing the complexity of the terminal.

[0285] The description of the specific examples in the embodiments can be referred to the corresponding description of the embodiments in FIGS. 1-6, which will not be repeated here.

[0286] The communication method related to the embodiments can include at least one of steps S501-S502. For example, step S501 can be implemented as an independent embodiment, and step S502 can be implemented as an independent embodiment. In addition, part or all of the steps S501-S502 can be combined to implement an independent embodiment, which is not limited in the embodiments.

[0287] The AI model determined according to the first information in the embodiments is applicable to multiple model input information and / or model output information, and can match different input and output dimensions and types corresponding to different AI use cases or different resource configurations.

[0288] The embodiments of the present disclosure also propose a device for implementing any of the above methods, for example, a device including units or modules for implementing each step performed by the terminal in any of the above methods. For another example, another device is also proposed, including units or modules for implementing each step performed by the network equipment (such as access network equipment, core network function node, core network equipment, etc.) in any of the above methods.

[0289] 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 the functions of any of the above methods or the units or modules 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 the 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 between the 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 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.

[0290] 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 hardware circuit, and the logical relationship of the hardware circuit is fixed or reconfigurable. 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, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), and the like.

[0291] FIG. 7 is a structural schematic diagram of a network device according to an embodiment of the present disclosure. As shown in FIG. 7, the network device can include a processing module 61. In some embodiments, the processing module 61 is configured to determine first information, and determine an AI model in the terminal according to the first information, the AI model including a first part, the first part being used to adapt to different model input information and / or model output information. Optionally, the processing module 61 is configured to perform at least one of the communication steps (for example, steps S301-S302, but not limited thereto) performed by the network device in any of the above methods, which will not be described herein again.

[0292] FIG. 8 is a schematic diagram of a structure of a terminal according to an embodiment of the present disclosure. As shown in FIG. 8, the terminal can include a processing module 71. In some embodiments, the processing module 71 is configured to determine first information, and determine an artificial intelligence (AI) model in the terminal according to the first information, the AI model including a first part for adapting to different model input information and / or model output information. Optionally, the processing module 71 is configured to perform at least one of the communication steps (for example, steps S401-S402, but not limited thereto) performed by the terminal in any of the above methods, which will not be described herein again.

[0293] In some embodiments, the processing module described above can be one module, or can include a plurality of sub-modules. Optionally, the plurality of sub-modules can perform all or part of the steps required to be performed by the processing module, respectively. Optionally, the processing module can be replaced by a processor.

[0294] FIG. 9 is a schematic diagram of a structure of a communication device 8100 according to an embodiment of the present disclosure. The communication device 8100 can be a network device (for example, an access network device, a core network device, etc.), a terminal (for example, a user equipment, etc.), a chip, a chip system, or a processor supporting the network device to implement any of the above methods, or a chip, a chip system, or a processor supporting the terminal to implement any of the above methods. The communication device 8100 can be used to implement the methods described in the above method embodiments, and details can be referred to the descriptions in the above method embodiments.

[0295] As shown in FIG. 9, the communication device 8100 includes one or more processors 8101. The processor 8101 can be a general-purpose processor or a special-purpose processor, for example, a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data, and the central processing unit can be used to control the communication device (for example, a base station, a baseband chip, a terminal device, a terminal device chip, a DU or a CU, etc.), execute programs, and process data of the programs. Optionally, the communication device 8100 is configured to perform any of the above methods. Optionally, the one or more processors 8101 are configured to invoke instructions to cause the communication device 8100 to perform any of the above methods.

[0296] In some embodiments, the communication device 8100 further includes one or more transceivers 8102. When the communication device 8100 includes one or more transceivers 8102, the transceiver 8102 performs at least one of the communication steps (for example, step S201, but not limited to) in the above-described method, and the processor 8101 performs at least one of the other steps (for example, steps S202, S203, but not limited to). In alternative embodiments, the transceiver can include a receiver and / or a transmitter, which can be separate or integrated together. Alternatively, the terms transceiver, transceiving unit, transceiver, transceiving circuit, interface circuit, interface, etc. can be replaced with each other, 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.

[0297] In some embodiments, the communication device 8100 further includes one or more memories 8103 for storing data. Alternatively, all or part of the memory 8103 can also be outside the communication device 8100. In alternative embodiments, the communication device 8100 can include one or more interface circuits 8104. Alternatively, the interface circuit 8104 is connected with the memory 8102, and the interface circuit 8104 can be used to receive data from the memory 8102 or other devices, and can be used to send data to the memory 8102 or other devices. For example, the interface circuit 8104 can read the data stored in the memory 8102 and send the data to the processor 8101.

[0298] The communication device 8100 described in the above embodiments can be a network device or a terminal, but the scope of the communication device 8100 described in the present disclosure is not limited thereto, and the structure of the communication device 8100 can not be limited by Figure 9. The communication device can be a standalone device or can be part of a larger device. For example, the communication device can be: 1) a standalone integrated circuit (IC), or a chip, or a chip system or subsystem; (2) a set of one or more ICs, which can optionally include 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, etc.; (6) others, etc.

[0299] Figure 10 is a structural schematic diagram of a chip 8200 according to an embodiment of the present disclosure. For the case where the communication device 8100 is a chip or a chip system, the structural schematic diagram of the chip 8200 shown in Figure 10 can be referred to, but is not limited thereto.

[0300] The chip 8200 includes one or more processors 8201. The chip 8200 is configured to perform any of the above methods.

[0301] In some embodiments, the chip 8200 further includes one or more interface circuits 8202. Optionally, the terms interface circuit, interface, transceiver pin, and the like can replace each other. In some embodiments, the chip 8200 further includes one or more memories 8203 for storing data. Optionally, all or part of the memory 8203 can be outside the chip 8200. Optionally, the interface circuit 8202 is connected with the memory 8203, the interface circuit 8202 can be configured to receive data from the memory 8203 or other devices, and the interface circuit 8202 can be configured to send data to the memory 8203 or other devices. For example, the interface circuit 8202 can read the data stored in the memory 8203 and send the data to the processor 8201.

[0302] In some embodiments, the interface circuit 8202 performs at least one of the communication steps such as sending and / or receiving in the above methods. The interface circuit 8202 performing the communication steps such as sending and / or receiving in the above methods means that the interface circuit 8202 performs data interaction between the processor 8201, the chip 8200, the memory 8203, or the transceiver device. In some embodiments, the processor 8201 performs at least one of the other steps.

[0303] The modules and / or devices described in each of the embodiments of the virtual device, the physical device, the chip, and the like can be combined or separated as appropriate. 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.

[0304] The disclosure also proposes a storage medium, and the above storage medium stores instructions, when the instructions run on the communication device 8100, the communication device 8100 performs any 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, 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, it can also be a transitory storage medium.

[0305] The disclosure also proposes a program product, and the above program product is executed by the communication device 8100, so that the communication device 8100 performs any of the above methods. Optionally, the above program product is a computer program product.

[0306] The disclosure also proposes a computer program, when it runs on a computer, the computer executes any of the above methods.

Claims

1. A communication method, characterized in that, Performed by a network device, the method includes: Determine the first piece of information; Based on the first information, an artificial intelligence (AI) model is determined in the terminal. The AI ​​model includes a first part, which is used to adapt to different model input information and / or model output information.

2. The method according to claim 1, characterized in that, The first part includes at least one of the following: An input adaptation section is used to adapt to different model input information; The output adaptation section is used to adapt to different model output information.

3. The method according to claim 2, characterized in that, The input adaptation part includes at least one first module, wherein the input parameter dimensions and / or input parameter types of the at least one first module are different, and the output parameter dimensions and / or output parameter types of the at least one first module are the same.

4. The method according to any one of claims 2 to 3, characterized in that, The output adaptation section includes at least one second module, wherein the input parameter dimensions and / or input parameter types of the at least one second module are the same, and the output parameter dimensions and / or output parameter types of the at least one second module are different.

5. The method according to any one of claims 2 to 4, characterized in that, The AI ​​model also includes a second part, which is used to learn the latent features in the model training set and the relationship between the latent features and the prediction target.

6. The method according to claim 5, characterized in that, The output parameters of the input adaptation part are used as the input parameters of the second part.

7. The method according to claim 6, characterized in that, The output parameter dimension of the input adaptation part is the same as the input parameter dimension of the second part, and / or the output parameter type of the input adaptation part is the same as the input parameter type of the second part.

8. The method according to any one of claims 5 to 7, characterized in that, The output parameters of the second part serve as the input parameters of the output adaptation part.

9. The method according to claim 8, characterized in that, The input parameter dimension of the output adaptation part is the same as the output parameter dimension of the second part, and / or the input parameter type of the output adaptation part is the same as the output parameter type of the second part.

10. The method according to any one of claims 5 to 9, characterized in that, The method further includes: Send the model parameters corresponding to the AI ​​model to the terminal; The model parameters include at least one of the following: The model parameters corresponding to the first part; The model parameters corresponding to the second part.

11. The method according to claim 10, characterized in that, The method further includes: Send the second information to the terminal; The second information is used to indicate the input adaptation part and / or output adaptation part corresponding to the model parameters.

12. The method according to claim 11, characterized in that, The second information includes at least one of the following: A first identifier is used to determine the input adaptation part corresponding to the model parameters; The second identifier is used to determine the output adaptation part corresponding to the model parameters.

13. The method according to any one of claims 1 to 12, characterized in that, The model input information includes at least one of the following: Model input dimensions; Model input type.

14. The method according to any one of claims 1 to 13, characterized in that, The model output information includes at least one of the following: Model output dimensions; Model output type.

15. The method according to any one of claims 1 to 14, characterized in that, The determination of the first information includes: Receive the first information sent by the terminal.

16. A communication method, characterized in that, The method, executed by a terminal, includes: Determine the first piece of information; Based on the first information, an artificial intelligence (AI) model is determined in the terminal. The AI ​​model includes a first part, which is used to adapt to different model input information and / or model output information.

17. The method according to claim 16, characterized in that, The first part includes at least one of the following: An input adaptation section is used to adapt to different model input information; The output adaptation section is used to adapt to different model output information.

18. The method according to claim 17, characterized in that, The input adaptation part includes at least one first module, wherein the input parameter dimensions and / or input parameter types of the at least one first module are different, and the output parameter dimensions and / or output parameter types of the at least one first module are the same.

19. The method according to any one of claims 17 to 18, characterized in that, The output adaptation section includes at least one second module, wherein the input parameter dimensions and / or input parameter types of the at least one second module are the same, and the output parameter dimensions and / or output parameter types of the at least one second module are different.

20. The method according to any one of claims 17 to 19, characterized in that, The AI ​​model also includes a second part, which is used to learn the latent features in the model training set and the relationship between the latent features and the prediction target.

21. The method according to claim 20, characterized in that, The output parameters of the input adaptation part are used as the input parameters of the second part.

22. The method according to claim 21, characterized in that, The output parameter dimension of the input adaptation part is the same as the input parameter dimension of the second part, and / or the output parameter type of the input adaptation part is the same as the input parameter type of the second part.

23. The method according to any one of claims 20 to 22, characterized in that, The output parameters of the second part serve as the input parameters of the output adaptation part.

24. The method according to claim 23, characterized in that, The input parameter dimension of the output adaptation part is the same as the output parameter dimension of the second part, and / or the input parameter type of the output adaptation part is the same as the output parameter type of the second part.

25. The method according to any one of claims 20 to 24, characterized in that, The method further includes: Receive model parameters corresponding to the AI ​​model sent by the network device; The model parameters include at least one of the following: The model parameters corresponding to the first part; The model parameters corresponding to the second part.

26. The method according to claim 25, characterized in that, The method further includes: Receive the second information sent by the network device; Based on the second information, determine the input adaptation part and / or output adaptation part corresponding to the model parameters.

27. The method according to claim 26, characterized in that, The second information includes at least one of the following: A first identifier is used to determine the input adaptation part corresponding to the model parameters; The second identifier is used to determine the output adaptation part corresponding to the model parameters.

28. The method according to any one of claims 16 to 27, characterized in that, The model input information includes at least one of the following: Model input dimensions; Model input type.

29. The method according to any one of claims 16 to 28, characterized in that, The model output information includes at least one of the following: Model output dimensions; Model output type.

30. The method according to any one of claims 16 to 29, characterized in that, The method further includes: Send the first information to the network device.

31. A communication method, characterized in that, include: The terminal sends the first message to the network device; The network device receives the first information sent by the terminal and determines the artificial intelligence (AI) model in the terminal based on the first information. The AI ​​model includes a first part, which is used to adapt to different model input information and / or model output information.

32. A network device, characterized in that, include: The processing module is configured to determine first information; and determine an artificial intelligence (AI) model in the terminal based on the first information, wherein the AI ​​model includes a first part, the first part being used to adapt to different model input information and / or model output information.

33. A terminal, characterized in that, include: The processing module is configured to determine first information; and determine an artificial intelligence (AI) model in the terminal based on the first information, wherein the AI ​​model includes a first part, the first part being used to adapt to different model input information and / or model output information.

34. A communication device, characterized in that, include: One or more processors; The processor is used to execute the method according to any one of claims 1 to 30.

35. A communication system, characterized in that, The invention includes a network device and a terminal, wherein the network device is configured to implement the method of any one of claims 1 to 15, and the terminal is configured to implement the method of any one of claims 16 to 30.

36. A computer storage medium, wherein, The computer storage medium stores computer-executable instructions; when executed by a processor, the computer-executable instructions can implement the method of any one of claims 1 to 30.

37. A computer program product comprising a computer program that, when executed by a processor, enables the implementation of the method according to any one of claims 1 to 30.

Images