Data transmission method, first device, second device, system, and storage medium

Abstract

The present disclosure provides a data transmission method, a first device, a second device, a system, and a storage medium. The method comprises: performing channel measurement on a reference signal sent by a third device to obtain channel measurement data; performing feature extraction and compression on the channel measurement data by means of a first artificial intelligence (AI) model to obtain first data; and sending the first data to the second device. The present disclosure can significantly reduce the transmission overhead of the channel measurement data used for AI-based positioning while ensuring positioning accuracy, thereby improving the practicality of the AI-based positioning.

Description

Data transmission method, first device, second device and system, and storage medium Technical Field

[0001] This disclosure relates to the field of communications, and in particular to data transmission methods, first devices, second devices and systems, and storage media. Background Technology

[0002] With the continuous development of artificial intelligence (AI) technology in recent years, AI-based solutions have been widely used in the field of wireless communication technology. Among them, high-precision positioning based on AI is an important application case of artificial intelligence technology in the field of communications.

[0003] Summary of the Invention

[0004] To improve the usability of AI positioning, embodiments of this disclosure provide a data transmission method, a first device, a second device and system, and a storage medium.

[0005] According to a first aspect of the present disclosure, a data transmission method is provided, the method being executed by a first device, comprising:

[0006] Channel measurement is performed on the reference signal sent by the third device to obtain channel measurement data;

[0007] The channel measurement data is feature-extracted and compressed using a first artificial intelligence (AI) model to obtain the first data.

[0008] The first data is sent to the second device.

[0009] According to a second aspect of the present disclosure, a data transmission method is provided, the method being executed by a second device, comprising:

[0010] Receive first data sent by a first device; wherein the first data is data obtained by the first device after extracting and compressing features from channel measurement data using a first artificial intelligence (AI) model, and the channel measurement data is data obtained by the first device after performing channel measurement on a reference signal sent by a third device;

[0011] Based on the first data and the second AI model, the location result is determined.

[0012] According to a third aspect of the present disclosure, a first device is provided, comprising:

[0013] The processing module is configured to perform channel measurements on the reference signal sent by the third device and obtain channel measurement data.

[0014] The processing module is further configured to extract and compress features from the channel measurement data using a first artificial intelligence (AI) model to obtain first data.

[0015] The transceiver module is configured to send the first data to the second device.

[0016] According to a fourth aspect of the present disclosure, a second device is provided, comprising:

[0017] The transceiver module is configured to receive first data sent by a first device; wherein the first data is data obtained by the first device after extracting and compressing features from channel measurement data using a first artificial intelligence (AI) model, and the channel measurement data is data obtained by the first device after performing channel measurements on a reference signal sent by a third device;

[0018] The processing module is configured to determine the location result based on the first data and the second AI model.

[0019] According to a fifth aspect of the present disclosure, a first device is provided, comprising:

[0020] One or more processors;

[0021] The processor is configured to execute the data transmission method described in any one of the first aspects.

[0022] According to a sixth aspect of the present disclosure, a second device is provided, comprising:

[0023] One or more processors;

[0024] The processor is used to execute the data transmission method described in any one of the second aspects.

[0025] According to a seventh aspect of the present disclosure, a communication system is provided, comprising:

[0026] A third device, configured to send a reference signal to the first device;

[0027] The first device is configured to implement the data transmission method described in any one of the first aspects;

[0028] The second device is configured to implement the data transmission method described in any one of the second aspects.

[0029] According to an eighth aspect of the present disclosure, a storage medium is provided that stores instructions that, when executed on a communication device, cause the communication device to perform a data transmission method as described in any one of the first or second aspects.

[0030] According to a ninth aspect of the present disclosure, a computer program product is provided, including a computer program that, when executed by a processor, is used to implement the data transmission method described in any one of the first or second aspects.

[0031] In this embodiment of the disclosure, the first device can extract and compress features from the channel measurement data using a first AI model, and send the obtained first data to the second device, thereby enabling the second device to locate the device. While ensuring the accuracy of the location, this significantly reduces the transmission overhead of the channel measurement data used for AI location and improves the availability of AI location.

[0032] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0033] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0034] Figure 1A is an exemplary schematic diagram of the architecture of a communication system provided according to an embodiment of the present disclosure.

[0035] Figure 1B is an exemplary scenario diagram of an AI-based positioning process provided according to an embodiment of the present disclosure.

[0036] Figure 1C is an exemplary scenario diagram of five AI-based positioning modes provided according to embodiments of the present disclosure.

[0037] Figure 2A is one of the exemplary interactive schematic diagrams of a data transmission method provided according to an embodiment of the present disclosure.

[0038] Figure 2B is a second exemplary interactive schematic diagram of a data transmission method provided according to an embodiment of the present disclosure.

[0039] Figure 3A is one of the exemplary flowcharts of a neural network model training method provided according to an embodiment of the present disclosure.

[0040] Figure 3B is a second exemplary flowchart of a neural network model training method provided according to an embodiment of the present disclosure.

[0041] Figure 4A is one of the exemplary flowcharts of a data transmission method provided according to an embodiment of the present disclosure.

[0042] Figure 4B is a second exemplary flowchart of a data transmission method provided according to an embodiment of the present disclosure.

[0043] Figure 5A is one of the exemplary scenarios of data transmission provided according to embodiments of the present disclosure.

[0044] Figure 5B is a third exemplary interactive schematic diagram of a data transmission method provided according to an embodiment of the present disclosure.

[0045] Figure 5C is a second exemplary scenario diagram of data transmission provided according to an embodiment of the present disclosure.

[0046] Figure 5D is a fourth exemplary interactive schematic diagram of a data transmission method provided according to an embodiment of the present disclosure.

[0047] Figure 6A is an exemplary block diagram of a first device provided according to an embodiment of the present disclosure.

[0048] Figure 6B is an exemplary block diagram of a second device provided according to an embodiment of the present disclosure.

[0049] Figure 7A is an exemplary block diagram of a communication device provided according to an embodiment of the present disclosure.

[0050] Figure 7B is an exemplary block diagram of a chip provided according to an embodiment of the present disclosure. Detailed Implementation

[0051] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0052] This disclosure provides a data transmission method, a first device, a second device and system, and a storage medium.

[0053] In a first aspect, embodiments of this disclosure propose a data transmission method, which is executed by a first device and includes: performing channel measurement on a reference signal sent by a third device to obtain channel measurement data; extracting and compressing features from the channel measurement data using a first artificial intelligence (AI) model to obtain first data; and sending the first data to a second device.

[0054] In the above embodiments, the first device can extract and compress features from the channel measurement data using the first AI model, and send the obtained first data to the second device, so that the second device can perform device positioning. While ensuring positioning accuracy, this can significantly reduce the transmission overhead of the channel measurement data used for AI positioning and improve the availability of AI positioning.

[0055] In conjunction with some embodiments of the first aspect, in some embodiments, the amount of the first data is less than the amount of the channel measurement data, and / or, the dimension corresponding to the first data is lower than the dimension corresponding to the channel measurement data.

[0056] In the above embodiments, the first device can reduce the amount of channel measurement data and / or reduce the dimensionality of channel measurement data through the first AI model, thereby reducing the transmission overhead of channel measurement data used for AI positioning and improving its availability.

[0057] In conjunction with some embodiments of the first aspect, in some embodiments, the training process of the first AI model and the second AI model includes the following steps: extracting and compressing features from channel measurement sample data using the first initial AI model to obtain estimated data; inputting the estimated data into the second initial AI model to obtain the estimated positioning result output by the second initial AI model; determining a first loss function based on the difference between the estimated positioning result and the true value of the positioning result; and training both the first initial AI model and the second initial AI model simultaneously based on the first loss function until a first stopping condition is met, thereby obtaining the first AI model and the second AI model.

[0058] In the above embodiments, the first initial AI model and the second initial AI model can be trained simultaneously to obtain the first AI model and the second AI model. While ensuring positioning accuracy, this can significantly reduce the transmission overhead of channel measurement data used for AI positioning and improve the availability of AI positioning.

[0059] In conjunction with some embodiments of the first aspect, in some embodiments, the first stopping condition includes at least one of the following: reaching a first number of training epochs; the first loss function decreasing to a first tolerance range; and the localization accuracy of the second initial AI model reaching a first value.

[0060] In the above embodiments, the first AI model and the second AI model can be obtained when the first stopping condition is met, which is simple to implement and highly usable.

[0061] In conjunction with some embodiments of the first aspect, in some embodiments, the training process of the first AI model and the third AI model includes the following steps: extracting and compressing features from channel measurement sample data using the first initial AI model to obtain estimated data; inputting the estimated data into the third initial AI model, which then restores the estimated data to obtain restored channel measurement sample data output by the third initial AI model; determining a second loss function based on the difference between the restored channel measurement sample data and the channel measurement sample data; and simultaneously training the first initial AI model and the third initial AI model based on the second loss function until a second stopping condition is met, thereby obtaining the first AI model and the third AI model.

[0062] In the above embodiments, the first initial AI model and the third initial AI model can be trained simultaneously to obtain the first AI model and the third AI model. While ensuring positioning accuracy, this can significantly reduce the transmission overhead of channel measurement data used for AI positioning and improve the availability of AI positioning.

[0063] In conjunction with some embodiments of the first aspect, in some embodiments, the second stopping condition includes at least one of the following: reaching a second number of training epochs; the second loss function decreasing to a second tolerance range; and the accuracy of the third initial AI model in data reconstruction reaching a second value.

[0064] In the above embodiments, the first AI model and the third AI model can be obtained when the first stopping condition is met, which is simple to implement and highly usable.

[0065] In conjunction with some embodiments of the first aspect, in some embodiments, the first device is a terminal, the third device is a network device, and the reference signal is a positioning reference signal (PRS); or the first device is a network device, the third device is a terminal, and the reference signal is a detection reference signal for positioning.

[0066] In the above embodiments, the reference signal can be PRS or SRS-Pos, which improves the reliability of AI positioning.

[0067] Secondly, this disclosure provides a data transmission method, which is executed by a second device and includes: receiving first data sent by a first device; wherein the first data is data obtained by the first device after feature extraction and compression of channel measurement data through a first artificial intelligence (AI) model, and the channel measurement data is data obtained by the first device after performing channel measurement on a reference signal sent by a third device; and determining a positioning result based on the first data and a second AI model.

[0068] In conjunction with some embodiments of the second aspect, in some embodiments, the amount of data in the first data is less than the amount of data in the channel measurement data, and / or the dimension corresponding to the first data is lower than the dimension corresponding to the channel measurement data.

[0069] In conjunction with some embodiments of the second aspect, in some embodiments, determining the positioning result based on the first data and the second AI model includes any one of the following: inputting the first data into the second AI model to obtain the positioning result output by the second AI model; or inputting the first data into the second AI model and determining the positioning result based on the intermediate positioning parameters output by the second AI model.

[0070] In conjunction with some embodiments of the second aspect, in some embodiments, the training process of the first AI model and the second AI model includes the following steps: extracting and compressing features from channel measurement sample data using the first initial AI model to obtain estimated data; inputting the estimated data into the second initial AI model to obtain the estimated positioning result output by the second initial AI model; determining a first loss function based on the difference between the estimated positioning result and the true positioning result; and training the first initial AI model and the second initial AI model simultaneously based on the first loss function until a first stopping condition is met, thereby obtaining the first AI model and the second AI model.

[0071] In conjunction with some embodiments of the second aspect, in some embodiments, the first stopping condition includes at least one of the following: reaching a first number of training epochs; the first loss function decreasing to a first tolerance range; and the localization accuracy of the second initial AI model reaching a first value.

[0072] In conjunction with some embodiments of the second aspect, in some embodiments, determining the positioning result based on the first data and the second AI model includes: inputting the first data into a third AI model to obtain the restored channel measurement data output by the third AI model; inputting the restored channel measurement data into the second AI model to obtain the positioning result output by the second AI model.

[0073] In conjunction with some embodiments of the second aspect, in some embodiments, determining the positioning result based on the first data and the second AI model includes: inputting the first data into a third AI model to obtain the restored channel measurement data output by the third AI model; inputting the restored channel measurement data into the second AI model to obtain intermediate positioning parameters output by the second AI model; and determining the positioning result based on the intermediate positioning parameters.

[0074] In conjunction with some embodiments of the second aspect, in some embodiments, the training process of the first AI model and the third AI model includes the following steps: extracting and compressing features from channel measurement sample data using the first initial AI model to obtain estimated data; inputting the estimated data into the third initial AI model to obtain the restored channel measurement sample data output by the third initial AI model; determining a second loss function based on the difference between the restored channel measurement sample data and the channel measurement sample data; and training both the first initial AI model and the third initial AI model simultaneously based on the second loss function until a second stopping condition is met, thereby obtaining the first AI model and the third AI model.

[0075] In conjunction with some embodiments of the second aspect, in some embodiments, the second stopping condition includes at least one of the following: reaching a second number of training epochs; the second loss function decreasing to a second tolerance range; and the accuracy of the third initial AI model in data reconstruction reaching a second value.

[0076] In conjunction with some embodiments of the second aspect, in some embodiments, the first device is a terminal, the third device is a network device, and the reference signal is a positioning reference signal (PRS); or the first device is a network device, the third device is a terminal, and the reference signal is a detection reference signal used for positioning.

[0077] Thirdly, embodiments of this disclosure propose a first device, comprising: a processing module configured to perform channel measurement on a reference signal transmitted by a third device to obtain channel measurement data; the processing module is further configured to perform feature extraction and compression on the channel measurement data using a first artificial intelligence (AI) model to obtain first data; and a transceiver module configured to transmit the first data to a second device.

[0078] Fourthly, this disclosure provides a second device, comprising: a transceiver module configured to receive first data sent by a first device; wherein the first data is data obtained by the first device after feature extraction and compression of channel measurement data through a first artificial intelligence (AI) model, and the channel measurement data is data obtained by the first device after performing channel measurement on a reference signal sent by a third device; and a processing module configured to determine a positioning result based on the first data and a second AI model.

[0079] Fifthly, embodiments of this disclosure provide a first device comprising: one or more processors; wherein the processors are configured to perform the data transmission method described in any one of the first aspects.

[0080] In a sixth aspect, embodiments of this disclosure provide a second device comprising: one or more processors; wherein the processors are configured to perform the data transmission method described in any one of the second aspects.

[0081] In a seventh aspect, embodiments of this disclosure provide a communication system comprising: a third device configured to transmit a reference signal to a first device; a first device configured to implement the data transmission method described in any one aspect; and a second device configured to implement the data transmission method described in any one aspect.

[0082] Eighthly, embodiments of this disclosure provide a storage medium storing instructions that, when executed on a communication device, cause the communication device to perform a data transmission method as described in any one of the first or second aspects.

[0083] In a ninth aspect, embodiments of this disclosure provide a computer program product, including a computer program that, when executed by a processor, is used to implement the data transmission method described in any one of the first or second aspects.

[0084] In a tenth aspect, embodiments of this disclosure provide a program product that, when executed by a communication device, causes the communication device to perform the method described in the optional implementation of the first aspect.

[0085] Eleventhly, embodiments of this disclosure provide a chip or chip system. The chip or chip system includes processing circuitry configured to perform the method described in the optional implementation of the first aspect above.

[0086] It is understood that the first device, the second device, the communication system, the storage medium, the program product, the computer program, the chip, or the chip system described above are all used to perform the methods proposed in the embodiments of this disclosure. Therefore, the beneficial effects that can be achieved can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0087] This disclosure provides a data transmission method, a first device, a second device and system, and a storage medium. In some embodiments, the terms "data transmission method" and "communication method," "device positioning method," and "data transmission device" can be used interchangeably; the terms "data transmission device" and "communication device," "device positioning device," and "data processing device" can be used interchangeably; and the terms "communication system," "data transmission system," and "device positioning system" can be used interchangeably.

[0088] This disclosure is not exhaustive, but merely illustrative of some embodiments, and is not intended to limit the scope of protection of this disclosure. Unless otherwise specified, each step in a particular embodiment can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, a solution after removing some steps in a particular embodiment can also be implemented as an independent embodiment, and the order of the steps in a particular embodiment can be arbitrarily interchanged. Furthermore, the optional implementation methods in a particular embodiment can be arbitrarily combined; moreover, the embodiments can be arbitrarily combined, for example, some or all steps of different embodiments can be arbitrarily combined, and a particular embodiment can be arbitrarily combined with the optional implementation methods of other embodiments.

[0089] In each of the disclosed embodiments, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of the embodiments are consistent and can be referenced by each other. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0090] The terminology used in the embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the scope of this disclosure.

[0091] In this embodiment of the disclosure, unless otherwise stated, elements expressed in the singular form, such as "a," "an," "the," "the," "the," "the," "the," "the," "this," etc., can mean "one and only one," or "one or more," "at least one," etc. For example, when using articles such as "a," "an," "the," etc. in translation, the noun following the article can be understood as either a singular expression or a plural expression.

[0092] In the embodiments disclosed herein, "multiple" refers to two or more.

[0093] In some embodiments, the terms “at least one of”, “one or more”, “a plurality of”, “multiple”, etc., may be used interchangeably.

[0094] In some embodiments, the notation "at least one of A and B", "A and / or B", "A in one case, B in another", "in response to one case A, in response to another case B", etc., may include the following technical solutions depending on the situation: in some embodiments, A (execute A regardless of B); in some embodiments, B (execute B regardless of A); in some embodiments, execution is selected from A and B (A and B are selectively executed); in some embodiments, A and B (both A and B are executed). The same applies when there are more branches such as A, B, C, etc.

[0095] In some embodiments, the notation "A or B" may include the following technical solutions, depending on the situation: in some embodiments, A (execution of A regardless of B); in some embodiments, B (execution of B regardless of A); in some embodiments, execution is selected from A and B (A and B are selectively executed). The same applies when there are more branches such as A, B, C, etc.

[0096] The prefixes "first," "second," etc., used in the embodiments of this disclosure are merely for distinguishing different descriptive objects and do not impose restrictions on the position, order, priority, quantity, or content of the descriptive objects. The description of the descriptive objects is found in the claims or the context of the embodiments, and the use of prefixes should not constitute unnecessary restrictions. For example, if the descriptive object is a "field," the ordinal numbers preceding "field" in "first field" and "second field" do not restrict the position or order of the "fields." "First" and "second" do not restrict whether the "fields" they modify are in the same message, nor do they restrict the order of "first field" and "second field." Similarly, if the descriptive object is a "level," the ordinal numbers preceding "level" in "first level" and "second level" do not restrict the priority between "levels." Furthermore, the number of descriptive objects is not limited by ordinal numbers and can be one or more. For example, in "first device," the number of "devices" can be one or more. Furthermore, the objects modified by different prefixes can be the same or different. For example, if the object being described is "device", then "first device" and "second device" can be the same device or different devices, and their types can be the same or different. Similarly, if the object being described is "information", then "first information" and "second information" can be the same information or different information, and their content can be the same or different.

[0097] In some embodiments, “including A,” “containing A,” “for indicating A,” and “carrying A” can be interpreted as directly carrying A or indirectly indicating A.

[0098] In some embodiments, devices, etc., can be interpreted as physical or virtual, and their names are not limited to the names recorded in the embodiments. Terms such as “device”, “equipment”, “circuit”, “network element”, “node”, “function”, “unit”, “section”, “system”, “network”, “chip”, “chip system”, “entity”, and “subject” can be used interchangeably.

[0099] In some embodiments, "device" may be interpreted as "network" or "terminal".

[0100] In some embodiments, "network" can be interpreted as devices included in a network (e.g., access network devices, core network devices, etc.).

[0101] In some embodiments, the terms "access network device (AN device)," "radio access network device (RAN device)," "base station (BS)," "radio base station," "fixed station," "node," "access point," "transmission point (TP)," "reception point (RP)," "transmission / reception point (TRP)," "panel," "antenna panel," "antenna array," "cell," "macro cell," "small cell," "femto cell," "pico cell," "sector," "cell group," "serving cell," "carrier," "component carrier," and "bandwidth part (BWP)" can be used interchangeably.

[0102] 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", and "client" can be used interchangeably.

[0103] In some embodiments, access network devices, core network devices, or network devices can be replaced by terminals. For example, embodiments of this disclosure can also be applied to structures where communication between access network devices, core network devices, or network devices and terminals is replaced by communication between multiple terminals (e.g., device-to-device (D2D), vehicle-to-everything (V2X), etc.). In this case, the structure can also be configured such that the terminal has all or part of the functions of the access network device. Furthermore, terms such as "uplink" and "downlink" can be replaced with terms corresponding to communication between terminals (e.g., "sidelink"). For example, uplink channel, downlink channel, etc., can be replaced with sidelink channel, and uplink link, downlink, etc., can be replaced with sidelink link.

[0104] In some embodiments, the terminal may be replaced by an access network device, a core network device, or a network device. In this case, the access network device, core network device, or network device may also be configured to have all or some of the functions of the terminal.

[0105] In some embodiments, the acquisition of data, information, etc., may comply with the laws and regulations of the country where the location is situated.

[0106] In some embodiments, data, information, etc., may be obtained with the user's consent.

[0107] Furthermore, each element, each row, or each column in the table of this disclosure can be implemented as an independent embodiment, and any combination of any element, any row, or any column can also be implemented as an independent embodiment.

[0108] Figure 1A is a schematic diagram of the architecture of a communication system according to an embodiment of the present disclosure.

[0109] As shown in Figure 1A, the communication system 100 includes a first device 101, a second device 102, and a third device 103.

[0110] In some embodiments, the first device 101 may be a device for receiving reference signals and performing channel measurements, such as a channel data measurement device.

[0111] In some embodiments, the first device 101 performs channel measurement on the reference signal sent by the third device 103 to obtain channel measurement data.

[0112] In some embodiments, channel measurement data may include, but is not limited to, channel impulse response (CIR).

[0113] For example, the name of the channel measurement data is not limited and can be interchanged with second data, data to be processed, etc.

[0114] In some embodiments, the first device 101 may be a terminal or a network device, such as an access network device or a core network device. In some embodiments, the terminal includes, but is not limited to, at least one of the following: mobile phone, wearable device, Internet of Things device, car with communication function, smart car, tablet computer, computer with wireless transceiver function, virtual reality (VR) terminal device, augmented reality (AR) terminal device, wireless terminal device in industrial control, wireless terminal device in self-driving, wireless terminal device in remote medical surgery, wireless terminal device in smart grid, wireless terminal device in transportation safety, wireless terminal device in smart city, and wireless terminal device in smart home.

[0115] In some embodiments, the access network device is, for example, a node or device that connects a terminal to a wireless network. The access network device may include, but is not limited to, at least one of the following in a 5G communication system: evolved Node B (eNB), next-generation eNB (ng-eNB), next-generation Node B (gNB), node B (NB), home node B (HNB), home evolved node B (HeNB), radio backhaul device, radio network controller (RNC), base station controller (BSC), base transceiver station (BTS), base band unit (BBU), mobile switching center, base station in a 6G communication system, open RAN, cloud RAN, base station in other communication systems, and access node in a Wi-Fi system.

[0116] In some embodiments, the first device 101 may be an autonomous driving device, an Internet of Things (IoT) device, an environmental IoT device, etc. Autonomous driving devices include, but are not limited to, driverless cars and drones. IoT devices include, but are not limited to, sensors using radio frequency identification (RFID), scanners, smart home devices, and robots. Environmental IoT devices include, but are not limited to, IoT devices that can obtain energy from the outside environment and be charged.

[0117] In some embodiments, the technical solutions of this disclosure can be applied to the Open RAN architecture. In this case, the interfaces between or within access network devices involved in the embodiments of this disclosure can be transformed into internal interfaces of Open RAN. The processes and information interactions between these internal interfaces can be implemented by software or programs.

[0118] In some embodiments, the access network device may be composed of a central unit (CU) and a distributed unit (DU). The CU may also be called a control unit. The CU-DU structure can separate the protocol layer of the access network device. Some of the protocol layer functions are centrally controlled by the CU, while the remaining part or all of the protocol layer functions are distributed in the DU and centrally controlled by the CU. However, this is not the only possibility.

[0119] In some embodiments, the core network equipment can be a single device, including a first network element, a second network element, etc., or it can be multiple devices or a group of devices, each including all or part of the first network element, the second network element, etc. Network elements can be virtual or physical. The core network includes, for example, at least one of the Evolved Packet Core (EPC), 5G Core Network (5GCN), and Next Generation Core (NGC).

[0120] In some embodiments, the second device 102 may be a device for locating the device using an AI model and determining the device location result, for example, the second device 102 may be an AI model inference device.

[0121] In some embodiments, at least a second AI model may be deployed on the second device 102 to perform device positioning. It is understood that device positioning can refer to locating a terminal, including but not limited to determining the terminal's location coordinates.

[0122] In some embodiments, the third device 103 may be a device for transmitting a reference signal, for example, the third device 103 may also be a reference signal transmitting device.

[0123] In some embodiments, the third device 103 may be a terminal, a network device, an autonomous driving device, an IoT device, an environmental IoT device, etc. Specific details will not be elaborated further here.

[0124] The above is merely an illustrative example. The solution provided in this disclosure can be described for the case where the first device 101 and the second device 102 are different devices.

[0125] It is understood that the communication system described in this disclosure is for the purpose of more clearly illustrating the technical solutions of this disclosure, and does not constitute a limitation on the technical solutions proposed in this disclosure. As those skilled in the art will know, with the evolution of system architecture and the emergence of new business scenarios, the technical solutions proposed in this disclosure are also applicable to similar technical problems.

[0126] The following embodiments of this disclosure can be applied to the communication system 100 shown in FIG1A, or to some of the main bodies, but are not limited thereto. The main bodies shown in FIG1A are illustrative. The communication system may include all or some of the main bodies in FIG1A, or it may include other main bodies outside of FIG1A. The number and form of each main body are arbitrary. Each main body may be physical or virtual. The connection relationship between the main bodies is illustrative. The main bodies may not be connected or may be connected. The connection can be in any way, it can be a direct connection or an indirect connection, it can be a wired connection or a wireless connection.

[0127] The embodiments disclosed herein can be applied to Long Term Evolution (LTE), LTE-Advanced (LTE-A), LTE-Beyond (LTE-B), SUPER 3G, IMT-Advanced, 4th generation mobile communication system (4G), 5th generation mobile communication system (5G), 5G new radio (NR), Future Radio Access (FRA), New-Radio Access Technology (RAT), New Radio (NR), New radio access (NX), Future generation radio access (FX), Global System for Mobile communications (GSM), CDMA2000, Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), and IEEE 802.20, Ultra-Wideband (UWB), Bluetooth (a registered trademark), Public Land Mobile Network (PLMN) networks, Device-to-Device (D2D) systems, Machine-to-Machine (M2M) systems, Internet of Things (IoT) systems, Vehicle-to-Everything (V2X) systems, systems utilizing other communication methods, and next-generation systems built upon them, etc. Furthermore, multiple systems can be combined (e.g., a combination of LTE or LTE-A with 5G).

[0128] In this disclosure, to meet the location service needs of various commercial service scenarios and industrial IoT scenarios, high-precision positioning technology has become a popular research direction, which helps to realize business services such as indoor navigation, augmented and virtual reality, and autonomous driving. Achieving high-precision positioning services based on existing wireless communication network infrastructure is an important direction in mobile communication technology research and a crucial part of the standardization research of wireless communication throughout history.

[0129] The standard introduces various positioning methods, including time-based and angle-based methods, to achieve high-precision positioning in both indoor and outdoor scenarios. With the increasing demand for related business services, a project to enhance positioning accuracy for commercial and Industrial Internet of Things (IIoT) scenarios has been initiated, aiming to achieve decimeter-level high-precision positioning and meet the high-precision location service needs of both consumer and enterprise markets. However, classic positioning algorithms, such as the Time Difference of Arrival (TDOA) and Multi-Round Trip Time (Multi-RTT) algorithms specified in the standard, struggle to meet the stringent positioning accuracy requirements of IIoT scenarios, thus impacting related business services.

[0130] With the continuous development of artificial intelligence (AI) technology in recent years, AI-based solutions have been widely applied in the field of wireless communication technology. Deep neural network models can effectively complete complex data processing and feature modeling processes and have been applied in wireless communication. Achieving high-precision positioning based on AI is an important application case of AI technology in the field of communication and one of the main research directions for standardization. Related technical solutions can utilize deep neural network models to model the mapping relationship between channel measurement data and terminal location coordinates, achieving higher positioning accuracy compared to traditional positioning methods.

[0131] Meanwhile, semantic communication technology based on artificial intelligence has flourished in recent years, providing a new research direction for future mobile communication data transmission. Compared with traditional communication technologies, semantic communication can extract key feature information from data, compress data while retaining important semantic information, thereby reducing data transmission overhead, and ensuring that the receiving end can complete data reconstruction with high accuracy, thus ensuring communication transmission quality. Implementing a semantic encoder at the sending end and a semantic decoder at the receiving end in a semantic communication system based on AI models is a common approach. This involves using neural network models to autonomously learn key semantic feature information from data based on large-scale training data, thereby completing data compression and reconstruction, achieving low-overhead and efficient data transmission.

[0132] In related technologies, research on AI-based high-precision positioning technology mainly considers two specific implementation methods, as shown in Figure 1B: direct positioning based on AI models and indirect positioning assisted by AI models.

[0133] (1) Direct positioning based on AI model: The input of the AI ​​positioning model is the channel measurement data used for positioning, and the output is the positioning result, i.e. the target location coordinates;

[0134] (2) AI model-assisted indirect positioning: The input of the AI ​​positioning model is channel measurement data used for positioning, and the output is intermediate positioning parameters, which may include time of arrival (ToA), angle of arrival (AoA), etc. Based on these intermediate parameters, traditional positioning methods such as TDOA can be used to calculate and obtain the positioning result, i.e., the target location coordinates.

[0135] For both direct and indirect AI-based positioning methods, the input to the AI ​​model is channel measurement data used for positioning, typically the CIR calculated based on a positioning reference signal. Specifically, the terminal can calculate the downlink CIR based on the positioning reference signal (PRS) transmitted by network devices, such as a base station, while the base station can calculate the uplink CIR based on the sounding reference signal-position (SRS-Pos) transmitted by the terminal for positioning.

[0136] Furthermore, depending on the specific device used for positioning, AI-based high-precision positioning can be implemented on terminals, base stations, or core network equipment, such as LMF. For example, the AI-based high-precision positioning scheme includes five specific application modes, as shown in Figure 1C.

[0137] Mode 1: Direct or indirect positioning on the user equipment (UE) side: The terminal calculates positioning measurements, such as CIR, based on the PRS sent by the base station (BS), and inputs this value into the AI ​​model to directly obtain the terminal's location coordinates. Alternatively, the terminal inputs the positioning measurements into the AI ​​model to obtain intermediate positioning parameters, such as ToA and AoA, and then uses traditional positioning methods such as TDOA to obtain the terminal's location coordinates. After obtaining the positioning result, i.e., the target location coordinates, the terminal reports the result to the LMF.

[0138] Mode 2, Terminal-Assisted LMF Side Indirect Positioning: The terminal calculates the measurement values ​​such as CIR for positioning based on the PRS sent by the base station, inputs them into the AI ​​model to obtain intermediate positioning parameters such as ToA and AoA, and reports the intermediate positioning parameters to the LMF. The LMF uses traditional positioning methods such as TDOA to obtain the terminal's location coordinates.

[0139] Mode 3, terminal-assisted direct positioning by the LMF: The terminal calculates the measurement value, such as CIR, for positioning based on the PRS sent by the base station, and reports the measurement value for positioning to the LMF. The LMF inputs the received measurement value into the AI ​​model to obtain the target location coordinates.

[0140] Mode 4, Terminal-Assisted LMF Side Indirect Positioning: The base station calculates and obtains positioning measurement values ​​such as CIR based on the SRS (SRS-Pos) sent by the terminal for positioning. The positioning measurement values ​​are input into the AI ​​model to obtain intermediate positioning parameters such as ToA and AoA. The intermediate positioning parameters are then reported to the LMF. The LMF uses traditional positioning methods such as TDOA to obtain the terminal's location coordinates.

[0141] Mode 5, base station-assisted direct positioning on the LMF side: The base station calculates and obtains positioning measurement values ​​such as CIR based on the SRS sent by the terminal for positioning, and reports the positioning measurement values ​​to the LMF. The LMF inputs the positioning measurement values ​​into the AI ​​model to directly obtain the terminal's location coordinates.

[0142] In the five modes described above, data acquisition and model inference may be implemented on different devices. For modes 1, 2, and 4, the channel measurement data required as input to the AI ​​positioning model is obtained by the terminal or base station based on positioning reference signals. The AI ​​model is also deployed on the terminal or base station. Therefore, the terminal or base station can directly input its acquired channel measurement data into the AI ​​positioning model to obtain the positioning result, without involving the reporting and transmission of channel measurement data used for positioning.

[0143] For modes 3 and 5, the channel measurement data used for AI positioning needs to be calculated and obtained by the terminal or base station based on the positioning reference signal, then reported to the LMF, input into the AI ​​positioning model, and the positioning result, i.e., the target location coordinates, is obtained.

[0144] In some embodiments, the main application process of the high-precision positioning scheme based on deep learning is as follows: after training the AI ​​network model for target positioning using training data, the trained AI positioning model is deployed on the terminal, base station or LMF to complete the target positioning work in the actual system, that is, to perform model inference.

[0145] For AI-based target localization applications in modes 3 and 5, the channel measurement data used for localization needs to be first reported by the terminal or base station to the LMF (Local Model Function), and then input into the AI ​​model to complete model inference and obtain the localization result. In related technologies, the CIR (Central Irregularity) calculated based on the localization reference signal is typically used as the input to the AI ​​model.

[0146] CIR can represent the effects of a signal propagating through a channel, reflecting the changes that occur after the signal has passed through the channel, including path loss and shadow fading causing the energy of the pulse signal to decrease, and the superposition of multiple different pulse signals received at the receiving end due to the existence of multiple propagation paths in the channel.

[0147] CIR data is a complex matrix whose dimensions are mainly related to the number of Transmission Reception Points (TRPs) and the number of time-domain sampling points. The CIR data of each TRP is a complex number on multiple time-domain sampling points, containing power, phase, and time delay information.

[0148] Because the raw CIR data has a large dimensionality, directly reporting the complete CIR data to the LMF by the terminal or base station would result in significant data transmission overhead, hindering the application of AI-based positioning technology in practical communication systems. CIR contains relatively complete channel state information and typically contains some information redundancy. By processing the CIR data and retaining only a portion of it, Power Delay Profile (PDP) and Delay Profile (DP) data can be obtained. Both types of data can also be used as input to AI positioning models.

[0149] PDP retains the power and delay information of CIR data. Its data dimensionality is related to the number of reference signal transmitting devices (TRPs) and the number of time-domain sampling points. The PDP data for each TRP consists of real numbers at multiple time-domain sampling points. Compared to CIR, PDP data has a lower dimensionality.

[0150] DP retains the time delay information of CIR data, and its data dimension is related to the number of reference signal transmitting devices (TRPs) and the number of time-domain sampling points.

[0151] Compared to CIR data, PDP and DP data have lower dimensionality. Using PDP or DP data to train and apply AI positioning models can effectively reduce the reporting overhead of channel measurement data used for positioning. In addition, according to the data distribution characteristics of CIR, there is less data with larger amplitudes containing more information in the time domain sampling point dimension. The data at most time domain sampling point locations is close to 0. Therefore, there is a large redundancy in transmitting complete data at each time domain sampling point. Applying specific data transmission methods to obtain channel measurement data with smaller data volume can also effectively reduce data reporting overhead.

[0152] However, the methods mentioned above for obtaining PDP, DP, and low-dimensional data using specific data processing techniques all require manual design and determination based on the actual data distribution characteristics. For example, methods might involve selecting data with larger amplitude values ​​or truncating data within a certain range. In some cases, using these data transmission methods may not conform to the actual data characteristics and could potentially lose certain key data feature information, thereby affecting the positioning accuracy of the AI ​​model.

[0153] To reduce the transmission overhead of channel measurement data used for AI positioning and improve the availability of AI positioning, this disclosure provides the following data transmission method, first device, second device and system, and storage medium.

[0154] Figure 2A is an interactive schematic diagram of a data transmission method according to an embodiment of the present disclosure. As shown in Figure 2A, the present disclosure relates to a data transmission method, which includes:

[0155] In step S2101, the third device 103 sends a reference signal to the first device 101.

[0156] In some embodiments, the third device 103 may be an access network device, such as a base station, the first device 101 may be a terminal, and the reference signal may be a PRS.

[0157] In some embodiments, the third device 103 is a terminal, the first device 101 can be an access network device, such as a base station, and the reference signal can be an SRS for positioning, i.e., SRS-Pos.

[0158] In some embodiments, the first device 101 receives a reference signal.

[0159] In step S2102, the first device 101 performs channel measurement on the reference signal to obtain channel measurement data.

[0160] In some embodiments, channel measurement data may include, but is not limited to, CIR.

[0161] In some embodiments, this disclosure does not limit the method by which the first device 101 obtains channel measurement data. Alternatively, other devices may perform channel measurements on the reference signal and then send the obtained channel measurement data to the first device 101.

[0162] In step S2103, the first device 101 extracts and compresses features from the channel measurement data using the first AI model to obtain the first data.

[0163] In some embodiments, the first AI model can be used to extract and compress features from the data.

[0164] In some embodiments, the first AI model is deployed on the first device 101.

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

[0166] In some embodiments, the first device 101 can input channel measurement data into the first AI model to obtain first data output by the first AI model.

[0167] In some embodiments, the amount of data in the first data is less than the amount of data in the channel measurement data, and / or the dimension corresponding to the first data is lower than the dimension corresponding to the channel measurement data.

[0168] In some embodiments, the name of the first AI model is not limited and can be interchanged with that of the semantic encoder.

[0169] In step S2104, the first device 101 sends the first data to the second device 102.

[0170] In some embodiments, the second device 102 may be a core network device, such as an LMF.

[0171] In some embodiments, the second device 102 receives the first data.

[0172] In step S2105, the second device 102 determines the positioning result based on the first data and the second AI model.

[0173] In some embodiments, a second AI model is deployed on the second device 102, and the second AI model is used for device localization. It is understood that device localization can refer to terminal localization, that is, determining the coordinate location of the terminal.

[0174] In some embodiments, the second AI model can be trained together with the first AI model. The specific training process will be described in subsequent embodiments and will not be described here.

[0175] In some embodiments, the second device 102 uses the first data as the input value of the second AI model to obtain the output value of the second AI model.

[0176] In one example, the output value could be a location result, such as the coordinates of the terminal.

[0177] In one example, the output value can be an intermediate positioning parameter, including but not limited to ToA, AoA, etc. If the output value is an intermediate positioning parameter, the second device 102 can use a traditional positioning method such as TDOA to calculate the positioning result, i.e., the position coordinates of the terminal, based on the intermediate positioning parameter.

[0178] In some embodiments, the names of signals, etc., are not limited to those described in the embodiments. Terms such as "information", "message", "signal", "signaling", "report", "configuration", "indication", "instruction", "command", "channel", "parameter", "domain", "field", "symbol", "symbol", "codebook", "codeword", "codepoint", "bit", "data", "program", and "chip" can be used interchangeably.

[0179] In some embodiments, “get,” “obtain,” “receive,” “transmit,” “bidirectional transmission,” and “send and / or receive” can be used interchangeably and can be interpreted as receiving from other entities, obtaining from protocols, obtaining from higher layers, obtaining through self-processing, or autonomous implementation, among other meanings.

[0180] In some embodiments, terms such as “send,” “transmit,” “report,” “distribute,” “transfer,” “bidirectional transmission,” “send and / or receive” can be used interchangeably.

[0181] In some embodiments, terms such as "certain," "preset," "default," "set," "indicated," "a certain," "any," and "first" can be used interchangeably. "Certain A," "preset A," "default A," "set A," "indicated A," "a certain A," "any A," and "first A" can be interpreted as A pre-defined in a protocol or the like, or as A obtained through setting, configuration, or instruction, or as specific A, a certain A, any A, or first A, but are not limited thereto.

[0182] The data transmission method involved in the embodiments of this disclosure may include at least one of steps S2101 to S2105. For example, step S2103 may be implemented as an independent embodiment, step S2103+S2104 may be implemented as an independent embodiment, step S2105 may be implemented as an independent embodiment, step S2103+S2104+S2105 may be implemented as an independent embodiment, step S2101 may be implemented as an independent embodiment, step S2102 may be implemented as an independent embodiment, step S2101+S2102 may be implemented as an independent embodiment, and steps S2101 to S2105 may be implemented as independent embodiments, but are not limited thereto.

[0183] In some embodiments, steps S2101 to S2105 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0184] In some embodiments, the execution order of steps S2101 to S2105 is not limited.

[0185] In the above embodiments, the first device can extract and compress features from the channel measurement data using the first AI model, and send the obtained first data to the second device, so that the second device can perform device positioning. While ensuring positioning accuracy, this can significantly reduce the transmission overhead of the channel measurement data used for AI positioning and improve the availability of AI positioning.

[0186] Figure 2B is an interactive schematic diagram of a data transmission method according to an embodiment of the present disclosure. As shown in Figure 2B, the present disclosure relates to a data transmission method, which includes:

[0187] In step S2201, the third device 103 sends a reference signal to the first device 101.

[0188] In some embodiments, step S2201 is implemented in a similar manner to step S2101 described above, and will not be repeated here.

[0189] In step S2202, the first device 101 performs channel measurement on the reference signal to obtain channel measurement data.

[0190] In some embodiments, step S2202 is implemented in a similar manner to step S2102 described above, and will not be repeated here.

[0191] In step S2203, the first device 101 extracts and compresses features from the channel measurement data using the first AI model to obtain the first data.

[0192] In some embodiments, step S2203 is implemented in a similar manner to step S2103 described above, and will not be repeated here.

[0193] In step S2204, the first device 101 sends the first data to the second device 102.

[0194] In some embodiments, step S2204 is implemented in a similar manner to step S2104 described above, and will not be repeated here.

[0195] In step S2205, the second device 102 determines the restored channel measurement data based on the first data.

[0196] In some embodiments, a third AI model is deployed on the second device 102.

[0197] In some embodiments, a third AI model can be used to reconstruct the data.

[0198] In one example, the second device 102 inputs the first data into the third AI model and obtains the restored channel measurement data output by the third AI model.

[0199] In some embodiments, the name of the third AI model is not limited and can be interchanged with that of the semantic decoder.

[0200] In some embodiments, the third AI model can be trained together with the first AI model. The specific training process will be described in subsequent embodiments and will not be described here.

[0201] In step S2206, the second device 102 determines the positioning result based on the restored channel measurement data and the second AI model.

[0202] In some embodiments, a second AI model is deployed on the second device 102, and the second AI model is used for device localization. It is understood that device localization can refer to terminal localization, that is, determining the coordinate location of the terminal.

[0203] In some embodiments, the second AI model can be trained independently. The specific training process will be described in subsequent embodiments and will not be described here.

[0204] In some embodiments, the second device 102 uses the restored channel measurement data as the input value of the second AI model to obtain the output value of the second AI model.

[0205] In one example, the output value could be a location result, such as the coordinates of the terminal.

[0206] In one example, the output value can be an intermediate positioning parameter, including but not limited to ToA, AoA, etc. If the output value is an intermediate positioning parameter, the second device 102 can use a traditional positioning method such as TDOA to calculate the positioning result, i.e., the position coordinates of the terminal, based on the intermediate positioning parameter.

[0207] In some embodiments, steps S2201 to S2206 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0208] In some embodiments, the execution order of steps S2201 to S2206 is not limited.

[0209] In the above embodiments, the first device can extract and compress features from the channel measurement data using the first AI model, and send the obtained first data to the second device, so that the second device can perform device positioning. While ensuring positioning accuracy, this can significantly reduce the transmission overhead of the channel measurement data used for AI positioning and improve the availability of AI positioning.

[0210] Figure 3A is a flowchart illustrating a neural network model training method according to an embodiment of the present disclosure. As shown in Figure 3A, the method can be executed by a first device 101, a second device 102, or other devices (e.g., a fourth device). The method involved in this embodiment of the present disclosure includes:

[0211] Step S3101: The channel measurement sample data is feature extracted and compressed using the first initial AI model to obtain the estimated data.

[0212] In some embodiments, the first initial AI model may use a Residual Network (ResNet), a Visual Geometry Group (VGG) network, or the like as its backbone network, and may include, but is not limited to, at least one of the following network layers: input layer; convolutional layer; pooling layer; activation function layer; connection layer; and output layer. This disclosure does not limit the architecture of the first initial AI model.

[0213] In some embodiments, channel measurement sample data can be input into a first initial AI model, which then performs feature extraction and compression on the channel measurement sample data to obtain the predicted data output by the first initial AI model.

[0214] Step S3102: Input the estimated data into the second initial AI model to obtain the estimated positioning result output by the second initial AI model.

[0215] In some embodiments, the second initial AI model may use ResNet, VGG network, or similar networks as its backbone network, and may include, but is not limited to, at least one of the following network layers: input layer; convolutional layer; pooling layer; activation function layer; connection layer; and output layer. This disclosure does not limit the architecture of the second initial AI model.

[0216] In some embodiments, the predicted data output by the first initial AI model can be used as the input value of the second initial AI model to obtain the predicted localization result output by the second initial AI model.

[0217] Step S3103: Determine the first loss function based on the difference between the estimated positioning result and the true positioning result.

[0218] In some embodiments, the first loss function may be determined based on the difference between the predicted positioning result and the true positioning result.

[0219] Understandably, the first and second initial AI models can be trained under supervision using the true values ​​of the localization results as labels.

[0220] The true value of the positioning result can be the actual coordinate position of the terminal corresponding to the channel measurement sample data.

[0221] Step S3104: Based on the first loss function, train the first initial AI model and the second initial AI model simultaneously until the first stopping condition is met, and then stop training to obtain the first AI model and the second AI model.

[0222] In some embodiments, the first stopping condition may include, but is not limited to, at least one of the following: reaching a first number of training epochs; the first loss function decreasing to a first tolerance range; and the localization accuracy of the second initial AI model reaching a first value.

[0223] In some embodiments, training can be stopped when a first stopping condition is met, at which point a first AI model and a second AI model are obtained.

[0224] Furthermore, the first AI model can be deployed on the first device 101, and the second AI model can be deployed on the second device 102.

[0225] In some embodiments, steps S3101 to S3104 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0226] In some embodiments, the execution order of steps S3101 to S3104 is not limited.

[0227] In the above embodiments, the first initial AI model and the second initial AI model can be trained simultaneously to obtain the first AI model and the second AI model. While ensuring positioning accuracy, this can significantly reduce the transmission overhead of channel measurement data used for AI positioning and improve the availability of AI positioning.

[0228] Figure 3B is a flowchart illustrating a neural network model training method according to an embodiment of the present disclosure. As shown in Figure 3B, the method can be executed by a first device 101, a second device 102, or other devices (e.g., a fourth device). The method involved in this embodiment of the present disclosure includes:

[0229] Step S3201: The channel measurement sample data is feature extracted and compressed using the first initial AI model to obtain the estimated data.

[0230] In some embodiments, step S3201 is implemented in a similar manner to step S3101 described above, and will not be repeated here.

[0231] Step S3202: Input the estimated data into the third initial AI model, and the third initial AI model restores the estimated data to obtain the restored channel measurement sample data output by the third initial AI model.

[0232] In some embodiments, the third initial AI model may use ResNet, VGG network, or similar networks as its backbone network, and may include, but is not limited to, at least one of the following network layers: input layer; convolutional layer; pooling layer; activation function layer; connection layer; and output layer. This disclosure does not limit the architecture of the third initial AI model.

[0233] In some embodiments, a third initial AI model is used for data reconstruction.

[0234] In some embodiments, the first data can be input into a third initial AI model, which then restores the estimated data to obtain the restored channel measurement sample data output by the third initial AI model.

[0235] Step S3203: Determine the second loss function based on the difference between the restored channel measurement sample data and the channel measurement sample data.

[0236] In some embodiments, the second loss function is determined based on the difference between the restored channel measurement sample data and the channel measurement sample data.

[0237] Step S3204: Based on the second loss function, train the first initial AI model and the third initial AI model simultaneously until the second stopping condition is met, and then stop training to obtain the first AI model and the third AI model.

[0238] In some embodiments, the second stopping condition includes, but is not limited to, at least one of the following: reaching a second number of training epochs; the second loss function decreasing to a second tolerance range; and the accuracy of the third initial AI model in data reconstruction reaching a second value.

[0239] In some embodiments, the first initial AI model and the third initial AI model are trained uniformly using channel measurement sample data as labels, and the first AI model and the third AI model are obtained when the second stopping condition is met.

[0240] Furthermore, the first AI model can be deployed on the first device 101, and the third AI model can be deployed on the second device 102.

[0241] In some embodiments, steps S3201 to S3204 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0242] In some embodiments, the execution order of steps S3201 to S3204 is not limited.

[0243] In the above embodiments, the first initial AI model and the third initial AI model can be trained simultaneously to obtain the first AI model and the third AI model. While ensuring positioning accuracy, this can significantly reduce the transmission overhead of channel measurement data used for AI positioning and improve the availability of AI positioning.

[0244] In some embodiments, the second AI model can be trained separately.

[0245] For example, channel measurement sample data can be input into a second initial AI model to obtain the predicted positioning result output by the second initial AI model. Based on the difference between the predicted positioning result and the true positioning result, a third loss function is determined. Based on the third loss function, the second initial AI model is trained, and training stops when a third stopping condition is met, thereby obtaining the second AI model.

[0246] The third stopping condition may include, but is not limited to, at least one of the following: reaching the third number of training cycles; the third loss function decreasing to the third fault tolerance range; and the localization accuracy of the second initial AI model reaching the first value.

[0247] Furthermore, the second AI model can be deployed on the second device 102.

[0248] In the above embodiments, the second initial AI model can be trained separately to obtain the second AI model. While ensuring positioning accuracy, it can significantly reduce the transmission overhead of channel measurement data used for AI positioning and improve the availability of AI positioning.

[0249] Figure 4A is a flowchart illustrating a data transmission method according to an embodiment of the present disclosure. As shown in Figure 4A, this embodiment relates to a data transmission method, which can be executed by a first device 101. The method includes:

[0250] Step S4101: Obtain channel measurement data.

[0251] In some embodiments, optional implementations of step S4101 can refer to optional implementations of step S2101 in FIG2A and other related parts in the embodiments involved in FIG2A, or can refer to optional implementations of step S2201 in FIG2B and other related parts in the embodiments involved in FIG2B, which will not be repeated here.

[0252] Step S4102: Obtain the first data.

[0253] In some embodiments, optional implementations of step S4102 can refer to optional implementations of step S2102 in FIG2A and other related parts in the embodiments involved in FIG2A, or can refer to optional implementations of step S2202 in FIG2B and other related parts in the embodiments involved in FIG2B, which will not be repeated here.

[0254] Step S4103: Send the first data.

[0255] In some embodiments, the first device 101 may send first data to the second device 102.

[0256] In some embodiments, the second device 102 receives the first data.

[0257] In some embodiments, optional implementations of step S4103 can refer to optional implementations of step S2103 in FIG2A and other related parts in the embodiments involved in FIG2A, or can refer to optional implementations of step S2203 in FIG2B and other related parts in the embodiments involved in FIG2B, which will not be repeated here.

[0258] In some embodiments, steps S4101 to S4103 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0259] In some embodiments, the execution order of steps S4101 to S4103 is not limited.

[0260] In the above embodiments, the first device can extract and compress features from the channel measurement data using the first AI model, and then send the obtained first data to the second device, which significantly reduces the transmission overhead of the channel measurement data used for AI positioning and improves the availability of AI positioning.

[0261] Figure 4B is a flowchart illustrating a data transmission method according to an embodiment of the present disclosure. As shown in Figure 4B, this embodiment of the present disclosure relates to a data transmission method, which can be executed by a second device 102. The method includes:

[0262] Step S4201: Obtain the first data.

[0263] In some embodiments, the second device 102 receives first data sent by the first device 101, but is not limited thereto, and may also receive first data sent by other entities.

[0264] In some embodiments, the second device 102 acquires first data as defined by a protocol.

[0265] In some embodiments, the second device 102 obtains first data from the upper layer(s).

[0266] In some embodiments, the second device 102 processes the data to obtain the first data.

[0267] In some embodiments, step S4201 is omitted, and the second device 102 autonomously implements the function indicated by the first data, or the above function is defaulted or set to default.

[0268] In some embodiments, optional implementations of step S4201 can refer to optional implementations of step S2104 in FIG2A and other related parts in the embodiments involved in FIG2A, or can refer to optional implementations of step S2204 in FIG2B and other related parts in the embodiments involved in FIG2B, which will not be repeated here.

[0269] Step S4202: Determine the positioning result.

[0270] In some embodiments, optional implementations of step S4202 can be found in optional implementations of step S2105 in FIG2A and other related parts in the embodiments involved in FIG2A, which will not be repeated here.

[0271] In some embodiments, optional implementations of step S4202 can be found in the optional implementations of steps S2205 to S2206 in FIG2B, and other related parts in the embodiments involved in FIG2B, which will not be repeated here.

[0272] In some embodiments, steps S4201 to S4202 are optional, and one or more of these steps may be omitted or substituted in different embodiments.

[0273] In some embodiments, the execution order of steps S4201 to S4202 is not limited.

[0274] In the above embodiments, the second device can obtain first data after feature extraction and compression from the first device, and determine the positioning result based on the first data and the second AI model. While ensuring positioning accuracy, it can significantly reduce the transmission overhead of channel measurement data used for AI positioning, thereby improving the usability of AI positioning.

[0275] The above process is further illustrated with examples below.

[0276] Since terminals or base stations need to report channel measurement data used for AI positioning to the LMF (Local Meter Function), directly reporting raw channel measurement data such as CIR (Channel Identifier) ​​would result in significant data reporting overhead. Furthermore, obtaining PDP (Programmable Pixel Filter), DP (Programmable Pixel Filter), or other data through fixed data transmission methods may sometimes fail to retain key feature information from the original data, thus affecting positioning accuracy. This invention proposes a channel measurement data processing and reporting method for AI positioning that combines semantic communication technology. Based on an AI semantic encoder, it completes feature extraction and compression of channel measurement data used for AI positioning. While ensuring positioning accuracy, it can significantly reduce the data transmission overhead of channel measurement data used for AI positioning, which is beneficial for promoting the practical application of AI-based target positioning technology in communication systems.

[0277] Example 1: In this embodiment of the present disclosure, an AI positioning data processing and reporting method combining semantic communication technology is proposed, mainly based on an AI semantic encoder, as shown in Figure 5A. After acquiring raw channel measurement data of types such as CIR based on positioning reference signals, the channel data measurement device (mainly including terminals or base stations) inputs the raw data into the AI ​​semantic encoder for feature extraction and data compression, obtaining semantic transmission data with a lower data volume that contains key semantic features from the raw data. This semantic data is then reported to an AI model inference device (mainly an LMF), which inputs the data into the AI ​​positioning model to obtain the positioning result.

[0278] The AI ​​semantic encoder is implemented based on a neural network model. Its input is raw channel measurement data, and its output is semantic transmission data. The output data dimension is lower than the input data dimension to achieve data feature extraction and dimensionality compression. The input data for the AI ​​localization model is semantic data after feature extraction and dimensionality compression. With reduced data dimensionality, the computational complexity of the AI ​​localization model is correspondingly reduced. During model training, the AI ​​semantic encoder and the AI ​​localization model are trained simultaneously. The output data of the AI ​​semantic encoder is directly input into the AI ​​localization model to calculate the localization result. The error between the actual output localization result and the labeled localization result in the training dataset is used as the loss function. The model is trained using methods such as stochastic gradient descent until the loss function decreases to a certain range, the training cycle reaches a certain number of iterations, or the model's localization accuracy reaches a certain requirement. After training, the AI ​​semantic encoder model will be deployed on channel data measurement equipment, and the AI ​​localization model will be deployed on AI model inference equipment.

[0279] The complete AI positioning model application process using the AI ​​positioning data processing and reporting method combining semantic communication technology proposed in this disclosure is shown in Figure 5B, including the following steps:

[0280] Step S5201: The reference signal transmitting device sends a reference signal for positioning to the channel data measurement device.

[0281] In application mode 3, the reference signal transmitting device is the base station, the positioning reference signal is the PRS, and the channel data measurement device is the terminal. In application mode 5, the reference signal transmitting device is the terminal, the positioning reference signal is the SRS used for positioning, i.e., SRS-Pos, and the channel data measurement device is the base station.

[0282] In step S5202, the channel data measurement device receives a reference signal for positioning and calculates and obtains channel measurement data, such as CIR, for AI positioning.

[0283] In step S5203, the channel data measurement device uses the AI ​​semantic encoder proposed in this embodiment to complete the feature extraction and compression of channel measurement data for AI positioning, and obtains semantic transmission data.

[0284] In step S5204, the channel data measurement device reports the semantic transmission data to the AI ​​model inference device.

[0285] In step S5205, the AI ​​model inference device acquires semantic reception data, inputs it into the AI ​​positioning model, and obtains the positioning result, i.e., the location coordinates of the positioning target.

[0286] Example 2: In this embodiment of the present disclosure, an AI positioning data processing and reporting method combining semantic communication technology is proposed, mainly based on an AI semantic encoder, as shown in Figure 5C. The channel data measurement device (mainly including a terminal or base station) acquires raw channel measurement data of types such as CIR based on the positioning reference signal, and then inputs it into the AI ​​semantic encoder for feature extraction and data compression. This yields semantic transmission data with a lower data volume that contains key semantic features from the raw data. The semantic data is then reported to the AI ​​model inference device (mainly an LMF). The AI ​​model inference device inputs the raw channel measurement data into the AI ​​semantic decoder to reconstruct the channel measurement data, and then inputs the reconstructed channel measurement data into the AI ​​positioning model to obtain the positioning result.

[0287] The AI ​​semantic encoder is implemented based on a neural network model. Its input is raw channel measurement data, and its output is semantic transmission data. The output data dimension is lower than the input data dimension to achieve data feature extraction and dimensionality compression. At this time, the input data of the AI ​​semantic decoder is semantic data after feature extraction and dimensionality compression. Its data dimension is reduced, and accordingly, the AI ​​semantic decoder can restore the original channel measurement data.

[0288] During model training, the AI ​​semantic encoder and AI semantic decoder are trained simultaneously. The output data of the AI ​​semantic encoder is directly input into the AI ​​semantic decoder. The error between the reconstructed channel measurement data and the original channel measurement data is used as the loss function. Model training is performed using methods such as stochastic gradient descent until the loss function decreases to a certain range, a certain number of training cycles are completed, or the model's localization accuracy reaches a certain requirement. After training, the AI ​​semantic encoder model will be deployed on channel data measurement equipment, and the AI ​​semantic decoder will be deployed on AI model inference equipment.

[0289] In addition, the AI ​​localization model can be trained separately. The specific training process has been described in the aforementioned embodiments and will not be repeated here.

[0290] The complete AI positioning model application process using the AI ​​positioning data processing and reporting method combining semantic communication technology proposed in this disclosure is shown in Figure 5D, including the following steps:

[0291] Step S5401: The reference signal transmitting device sends a reference signal for positioning to the channel data measurement device.

[0292] In application mode 3, the reference signal transmitting device is the base station, the positioning reference signal is the PRS, and the channel data measurement device is the terminal. In application mode 5, the reference signal transmitting device is the terminal, the positioning reference signal is the SRS used for positioning, and the channel data measurement device is the base station.

[0293] In step S5402, the channel data measurement device receives a reference signal for positioning and calculates and obtains channel measurement data, such as CIR, for AI positioning.

[0294] In step S5403, the channel data measurement device uses the AI ​​semantic encoder proposed in this embodiment to complete the feature extraction and compression of channel measurement data for AI positioning, and obtains semantic transmission data.

[0295] In step S5404, the channel data measurement device reports the semantic transmission data to the AI ​​model inference device.

[0296] Step S5405: The AI ​​model inference device acquires the restored channel data measurement.

[0297] Step S5406: Input the restored channel measurement data into the AI ​​positioning model to obtain the positioning result, i.e., the location coordinates of the positioning target.

[0298] In the embodiments disclosed herein, some or all of the steps and their optional implementations may be arbitrarily combined with some or all of the steps in other embodiments, or may be arbitrarily combined with the optional implementations in other embodiments.

[0299] This disclosure also proposes an apparatus for implementing any of the above methods. For example, an apparatus is proposed that includes units or modules for implementing the steps performed by the first device in any of the above methods.

[0300] It should be understood that the division of units or modules in the above device is only a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, the units or modules in the device can be implemented by a processor calling software: for example, the device includes a processor connected to a memory containing instructions. The processor calls the instructions stored in the memory to implement any of the above methods or to implement the functions of the units or modules in the above device. The processor can be, for example, a general-purpose processor, such as a Central Processing Unit (CPU) or a microprocessor, and the memory can be internal or external to the device. Alternatively, the units or modules in the device can be implemented in the form of hardware circuits. The functionality of some or all of the units or modules can be achieved through the design of these hardware circuits, which can be understood as one or more processors. For example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC). The functionality of some or all of the units or modules is achieved through the design of the logical relationships between the components within the circuit. In another implementation, the hardware circuit can be implemented using a programmable logic device (PLD). Taking a field-programmable gate array (FPGA) as an example, it can include a large number of logic gates. The connection relationships between the logic gates are configured through configuration files, thereby achieving the functionality of some or all of the units or modules. All units or modules of the above device can be implemented entirely through processor-called software, entirely through hardware circuits, or partially through processor-called software with the remaining parts implemented through hardware circuits.

[0301] In this embodiment, the processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction read and execute capabilities, such as a Central Processing Unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a microprocessor), or a digital signal processor (DSP). In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. The logical relationships of the aforementioned hardware circuits are fixed or reconfigurable. For example, the processor is a hardware circuit implemented using an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document and configuring the hardware circuit can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units or modules. Furthermore, 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), or a Deep Learning Processing Unit (DPU).

[0302] Figure 6A is a schematic diagram of the structure of the first device proposed in an embodiment of this disclosure. As shown in Figure 6A, the first device 6100 may include: a processing module 6101 and a transceiver module 6102.

[0303] In some embodiments, the processing module 6101 is configured to perform channel measurement on a reference signal sent by a third device to obtain channel measurement data; and to extract and compress features from the channel measurement data using a first artificial intelligence (AI) model to obtain first data.

[0304] In some embodiments, the transceiver module 6102 is configured to send the first data to a second device.

[0305] Optionally, the processing module 6101 is used to execute at least one of the other steps (such as steps S2102, S2103, S2202, and S2203, but not limited thereto) executed by the first device 6100 in any of the above methods, which will not be described in detail here.

[0306] Optionally, the transceiver module 6102 is used to perform at least one of the communication steps such as sending and / or receiving performed by the first device 6100 in any of the above methods (e.g., steps S2101, S2104, S2201, S2204, but not limited thereto), which will not be elaborated here.

[0307] Figure 6B is a schematic diagram of the structure of the first device proposed in an embodiment of this disclosure. As shown in Figure 6B, the second device 6200 may include: a transceiver module 6201 and a processing module 6202.

[0308] In some embodiments, the transceiver module 6201 is configured to receive first data sent by the first device; wherein the first data is data obtained by the first device after performing feature extraction and compression on channel measurement data through a first artificial intelligence (AI) model, and the channel measurement data is data obtained by the first device after performing channel measurement on a reference signal sent by the third device.

[0309] In some embodiments, the processing module 6202 is configured to determine the positioning result based on the first data and the second AI model.

[0310] Optionally, the transceiver module 6201 is used to perform at least one of the communication steps such as sending and / or receiving performed by the second device 6200 in any of the above methods (e.g., step S2104, step S2204, but not limited thereto), which will not be described in detail here.

[0311] Optionally, the processing module 6202 is used to execute at least one of the other steps (such as step S2105, step S2205, step S2206, but not limited thereto) executed by the second device 6200 in any of the above methods, which will not be described in detail here.

[0312] In some embodiments, the processing module may be a single module or may include multiple sub-modules. Optionally, the multiple sub-modules may each perform all or part of the steps required by the processing module. Optionally, the processing module may be interchangeable with a processor.

[0313] In some embodiments, the transceiver module may include a transmitting module and / or a receiving module, which may be separate or integrated. Optionally, the transceiver module may be interchangeable with a transceiver.

[0314] Figure 7A is a schematic diagram of the structure of the communication device 7100 proposed in an embodiment of this disclosure. The communication device 7100 can be a first device or a second device. The first device can be a network device (e.g., access network device, core network device, etc.), a terminal (e.g., user equipment, vehicle, etc.), or a chip, chip system, or processor that supports the network device in implementing any of the above methods, or a chip, chip system, or processor that supports the terminal in implementing any of the above methods. The second device can be a core network device, such as an LMF, or a chip, chip system, or processor that supports the core network device in implementing any of the above methods, or a chip, chip system, or processor that supports the core network device in implementing any of the above methods. The communication device 7100 can be used to implement the methods described in the above method embodiments; for details, please refer to the descriptions in the above method embodiments.

[0315] As shown in Figure 7A, the communication device 7100 includes one or more processors 7101. The processor 7101 can be a general-purpose processor or a dedicated processor, such as a baseband processor or a central processing unit (CPU). The baseband processor can be used to process communication protocols and communication data, while the CPU can be used to control communication devices (e.g., base stations, baseband chips, terminal devices, terminal device chips, DUs or CUs, etc.), execute programs, and process program data. Optionally, the communication device 7100 can be used to execute any of the above methods. Optionally, one or more processors 7101 can be used to invoke instructions to cause the communication device 7100 to execute any of the above methods.

[0316] In some embodiments, the communication device 7100 further includes one or more transceivers 7102. When the communication device 7100 includes one or more transceivers 7102, the transceivers 7102 perform at least one of the communication steps such as sending and / or receiving in the above method (e.g., steps S2101, S2104, S2201, S2204, but not limited thereto), and the processor 7101 performs at least one of other steps (e.g., steps S2102, S2103, S2105, S2202, S2203, S2205, S2206, but not limited thereto). In optional embodiments, the transceivers may include a receiver and / or a transmitter, which may be separate or integrated together. Optionally, terms such as transceiver, transceiver unit, transceiver, transceiver circuit, interface circuit, and interface can be used interchangeably; terms such as transmitter, transmitting unit, transmitter, and transmitting circuit can be used interchangeably; and terms such as receiver, receiving unit, receiver, and receiving circuit can be used interchangeably.

[0317] In some embodiments, the communication device 7100 further includes one or more memories 7103 for storing data. Optionally, all or part of the memories 7103 may be located outside the communication device 7100. In optional embodiments, the communication device 7100 may include one or more interface circuits 7104. Optionally, the interface circuits 7104 are connected to the memories 7103 and can be used to receive data from the memories 7103 or other devices, and to send data to the memories 7103 or other devices. For example, the interface circuits 7104 can read data stored in the memories 7103 and send the data to the processor 7101.

[0318] The communication device 7100 described in the above embodiments may be a network device or a terminal, but the scope of the communication device 7100 described in this disclosure is not limited thereto, and the structure of the communication device 7100 may not be limited by FIG. 7A. The communication device may be a standalone device or a part of a larger device. For example, the communication device may be: (1) a standalone integrated circuit IC, or chip, or chip system or subsystem; (2) a collection of one or more ICs, optionally, the IC collection may also include storage components for storing data and programs; (3) an ASIC, such as a modem; (4) a module that can be embedded in other devices; (5) a receiver, terminal device, smart terminal device, cellular phone, wireless device, handheld device, mobile unit, vehicle device, network device, cloud device, artificial intelligence device, etc.; (6) others, etc.

[0319] Figure 7B is a schematic diagram of the structure of the chip 7200 according to an embodiment of this disclosure. For cases where the communication device 7100 can be a chip or a chip system, the schematic diagram of the chip 7200 shown in Figure 7B can be referenced, but is not limited thereto.

[0320] Chip 7200 includes one or more processors 7201. Chip 7200 is used to perform any of the above methods.

[0321] In some embodiments, chip 7200 further includes one or more interface circuits 7202. Optionally, terms such as interface circuit, interface, and transceiver pin can be used interchangeably. In some embodiments, chip 7200 further includes one or more memories 7203 for storing data. Optionally, all or part of the memories 7203 may be located outside chip 7200. Optionally, interface circuit 7202 is connected to memory 7203, and interface circuit 7202 can be used to receive data from memory 7203 or other devices, and interface circuit 7202 can be used to send data to memory 7203 or other devices. For example, interface circuit 7202 can read data stored in memory 7203 and send the data to processor 7201.

[0322] In some embodiments, the interface circuit 7202 performs at least one of the communication steps such as sending and / or receiving in the above-described method (e.g., steps S2101, S2104, S2201, and S2204, but not limited thereto). The interface circuit 7202 performing the communication steps such as sending and / or receiving in the above-described method refers, for example, to the interface circuit 7202 performing data interaction between the processor 7201, the chip 7200, the memory 7203, or the transceiver device. In some embodiments, the processor 7201 performs at least one of other steps (e.g., steps S2102, S2103, S2105, S2202, S2203, S2205, and S2206, but not limited thereto).

[0323] The modules and / or devices described in the various embodiments, such as virtual devices, physical devices, and chips, can be combined or separated arbitrarily as needed. Optionally, some or all steps can also be performed collaboratively by multiple modules and / or devices, which is not limited here.

[0324] This disclosure also proposes a storage medium storing instructions that, when executed on the communication device 7100, cause the communication device 7100 to perform any of the above methods. Optionally, the storage medium is an electronic storage medium. Optionally, the storage medium is a computer-readable storage medium, but not limited thereto; it may also be a storage medium readable by other devices. Optionally, the storage medium may be a non-transitory storage medium, but not limited thereto; it may also be a temporary storage medium.

[0325] This disclosure also provides a program product that, when executed by the communication device 7100, causes the communication device 7100 to perform any of the above methods. Optionally, the program product is a computer program product.

[0326] This disclosure also proposes a computer program that, when run on a computer, causes the computer to perform any of the above methods.

[0327] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A data transmission method, characterized by, The method is performed by a first device, comprising: performing channel measurement on a reference signal transmitted by a third device to obtain channel measurement data; performing feature extraction and compression on the channel measurement data by a first artificial intelligence (AI) model to obtain first data; transmitting the first data to a second device.

2. The method of claim 1, wherein, The data amount of the first data is less than that of the channel measurement data, and / or the dimension corresponding to the first data is lower than that corresponding to the channel measurement data.

3. The method according to claim 1 or 2, characterized in that, The training process of the first AI model and a second AI model comprises the following steps: performing feature extraction and compression on channel measurement sample data by a first initial AI model to obtain estimated data; inputting the estimated data into a second initial AI model to obtain an estimated positioning result output by the second initial AI model; determining a first loss function based on the difference between the estimated positioning result and a true value of the positioning result; training the first initial AI model and the second initial AI model based on the first loss function until the training is stopped when a first stop condition is met, to obtain the first AI model and the second AI model.

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

5. The method according to claim 1 or 2, characterized in that, The training process of the first AI model and a third AI model comprises the following steps: performing feature extraction and compression on channel measurement sample data by a first initial AI model to obtain estimated data; inputting the estimated data into a third initial AI model to restore the estimated data by the third initial AI model, to obtain restored channel measurement sample data output by the third initial AI model; determining a second loss function based on the difference between the restored channel measurement sample data and the channel measurement sample data; training the first initial AI model and the third initial AI model based on the second loss function until the training is stopped when a second stop condition is met, to obtain the first AI model and the third AI model.

6. The method of claim 5, wherein, The second stop condition comprises at least one of the following: a second training cycle number is reached; the second loss function is reduced to within a second fault tolerance range; the accuracy of data restoration by the third initial AI model reaches a second value.

7. The method according to any one of claims 1 to 6, characterized in that, The first device is a terminal, the third device is a network device, and the reference signal is a positioning reference signal (PRS); or The first device is a network device, the third device is a terminal, and the reference signal is a sounding reference signal for positioning.

8. A data transmission method, characterized by, The method is performed by a second device, comprising: receiving first data transmitted by a first device; wherein the first data is obtained by the first device by performing feature extraction and compression on channel measurement data by a first artificial intelligence (AI) model, and the channel measurement data is obtained by the first device by performing channel measurement on a reference signal transmitted by a third device; determining a positioning result based on the first data and a second AI model.

9. The method of claim 8, wherein, The data amount of the first data is less than that of the channel measurement data, and / or the dimension corresponding to the first data is lower than that corresponding to the channel measurement data.

10. The method according to claim 8 or 9, characterized in that, The determining of the positioning result based on the first data and the second AI model comprises any one of the following: inputting the first data into the second AI model to obtain the positioning result output by the second AI model; inputting the first data into the second AI model, and determining the positioning result based on an intermediate positioning parameter output by the second AI model.

11. The method of claim 10, wherein, The training process of the first AI model and the second AI model comprises the following steps: performing feature extraction and compression on channel measurement sample data by a first initial AI model to obtain estimated data; inputting the estimated data into a second initial AI model to obtain estimated positioning result output by the second initial AI model; determining a first loss function based on the difference between the estimated positioning result and a true value of the positioning result; training the first initial AI model and the second initial AI model based on the first loss function until stopping training when a first stop condition is met, to obtain the first AI model and the second AI model.

12. The method of claim 11, wherein, The first stop condition comprises at least one of the following: a first training cycle number is reached; the first loss function falls within a first fault tolerance range; the positioning accuracy of the second initial AI model reaches a first value.

13. The method of claim 8 or 9, wherein, The determining of the positioning result based on the first data and the second AI model comprises: inputting the first data into a third AI model to obtain restored channel measurement data output by the third AI model; inputting the restored channel measurement data into the second AI model to obtain the positioning result output by the second AI model.

14. The method of claim 8 or 9, wherein, The determining of the positioning result based on the first data and the second AI model comprises: inputting the first data into a third AI model to obtain restored channel measurement data output by the third AI model; inputting the restored channel measurement data into the second AI model to obtain an intermediate positioning parameter output by the second AI model; determining the positioning result based on the intermediate positioning parameter.

15. The method according to claim 13 or 14, characterized in that, The training process of the first AI model and the third AI model comprises the following steps: performing feature extraction and compression on channel measurement sample data by a first initial AI model to obtain estimated data; inputting the estimated data into a third initial AI model to obtain restored channel measurement sample data output by the third initial AI model; determining a second loss function based on the difference between the restored channel measurement sample data and the channel measurement sample data; training the first initial AI model and the third initial AI model based on the second loss function until stopping training when a second stop condition is met, to obtain the first AI model and the third AI model.

16. The method of claim 15, wherein, The second stop condition comprises at least one of the following: a second training cycle number is reached; the second loss function falls within a second fault tolerance range; the accuracy of data restoration by the third initial AI model reaches a second value.

17. The method according to any one of claims 8-16, characterized in that, The first device is a terminal, the third device is a network device, and the reference signal is a positioning reference signal (PRS); or The first device is a network device, the third device is a terminal, and the reference signal is a sounding reference signal (SRS) for positioning.

18. A first device, comprising: Comprising: a processing module configured to perform channel measurement on a reference signal transmitted by a third device to obtain channel measurement data; The processing module is further configured to perform feature extraction and compression on the channel measurement data by a first artificial intelligence (AI) model to obtain first data; a transceiver module configured to transmit the first data to a second device.

19. A second device, comprising: Comprising: a transceiver module configured to receive first data transmitted by a first device; wherein the first data is data obtained by the first device by performing feature extraction and compression on channel measurement data by a first artificial intelligence (AI) model, and the channel measurement data is data obtained by the first device by performing channel measurement on a reference signal transmitted by a third device; a processing module configured to determine a positioning result based on the first data and a second AI model.

20. A first device, comprising: Comprising: one or more processors; wherein the processor is configured to perform the data transmission method of any one of claims 1-7.

21. A second device, comprising: Comprising: one or more processors; wherein the processor is configured to perform the data transmission method of any one of claims 8-17.

22. A communication system, characterized by Comprising: a third device configured to transmit a reference signal to a first device; the first device, which is configured to implement the data transmission method of any one of claims 1-7; a second device configured to implement the data transmission method of any one of claims 8-17.

23. A storage medium, the storage medium storing instructions, wherein, When the instructions are run on a communication device, the communication device is caused to perform the data transmission method of any one of claims 1-7 or 8-17.

24. A computer program product comprising a computer program, characterized in that, The computer program is executed by a processor to implement the data transmission method of any one of claims 1-7 or 8-17.

Images