WO2023185865A1 - Model validation feedback method and apparatus, terminal, and network side device - Google Patents

Abstract

The present application relates to the field of mobile communications, and discloses a model validation feedback method and apparatus, a terminal, and a network side device. The model validation feedback method in the embodiments of the present application comprises: a first communication device obtains first measurement information of N target terminals on the basis of a first model; and the first communication device determines target information according to the first measurement information of the N target terminals, wherein the target information comprises validation information of relevant information of the first model, and N is a positive integer.

Description

Model validity feedback method, device, terminal and network side equipment

cross reference

This invention requires the priority of a Chinese patent application submitted to the China Patent Office on March 29, 2022, with the application number 202210320881.X and the invention name "Model Validity Feedback Method, Device, Terminal and Network Side Equipment". The entire contents of are incorporated herein by reference.

Technical field

This application belongs to the field of mobile communication technology, and specifically relates to a model effectiveness feedback method, device, terminal and network side equipment.

Background technique

Artificial Intelligence (AI) is currently widely used in various fields. Taking AI-based positioning methods as an example, positioning services can be assisted by pre-built and trained AI models.

However, as the measurement environment and requirements for positioning services vary, the data and requirements will change over time. There is no guarantee that the AI model will still be effective during long-term use or in different scenarios.

Contents of the invention

Embodiments of the present application provide a model effectiveness feedback method, device, terminal and network side equipment, which can solve the problem of being unable to guarantee that the AI model is still effective during long-term use or in different scenarios.

In a first aspect, a model validity feedback method is provided, applied to the first communication device, and the method includes:

The first communication device obtains first measurement information of N target terminals based on the first model;

The first communication device determines target information based on the first measurement information of the N target terminals;

Wherein, the target information includes validity information of the related information of the first model, and N is a positive integer.

In the second aspect, a model effectiveness feedback device is provided, including:

A model inference module, used to obtain the first measurement information of N target terminals based on the first model;

A validity judgment module, configured for the first communication device to determine target information based on the first measurement information of the N target terminals;

Wherein, the target information includes validity information of the related information of the first model, and N is a positive integer.

In a third aspect, a terminal is provided. The terminal includes a processor and a memory. The memory stores programs or instructions that can be run on the processor. When the program or instructions are executed by the processor, the following implementations are implemented: The steps of the method described in one aspect.

In a fourth aspect, a terminal is provided, including a processor and a communication interface, wherein the processor is configured to obtain first measurement information of N target terminals based on a first model; according to the first measurement information of the N target terminals The information determines the target information, and the communication interface is used to send feedback information.

In a fifth aspect, a network side device is provided. The network side device includes a processor and a memory. The memory stores programs or instructions that can be run on the processor. The program or instructions are executed by the processor. When implementing the steps of the method described in the first aspect.

In a sixth aspect, a network side device is provided, including a processor and a communication interface, wherein the processor is configured to obtain first measurement information of N target terminals based on a first model; according to the first measurement information of the N target terminals, A measurement information determines target information, and the communication interface is used to send feedback information.

In a seventh aspect, a model validity feedback system is provided, including: a terminal and a network side device. The terminal can be used to perform the steps of the model validity feedback method as described in the first aspect. The network side device can be used to The steps of the model effectiveness feedback method as described in the first aspect are performed.

In an eighth aspect, a readable storage medium is provided, the readable storage medium stores a program or Instructions, programs or instructions when executed by a processor implement the steps of the method described in the first aspect.

In a ninth aspect, a chip is provided. The chip includes a processor and a communication interface. The communication interface is coupled to the processor. The processor is used to run programs or instructions to implement the method described in the first aspect. .

In a tenth aspect, a computer program/program product is provided, the computer program/program product is stored in a storage medium, and the computer program/program product is executed by at least one processor to implement the method as described in the first aspect Steps of the model effectiveness feedback method.

In this embodiment of the present application, the first measurement information of N target terminals is obtained based on the first model through the first communication device; the first communication device determines the target information based on the first measurement information of the N target terminals, so that The effectiveness of the first model can be accurately obtained, and the first model can be monitored in real time to ensure that the first model can meet actual application requirements.

Description of drawings

Figure 1 is a schematic structural diagram of a wireless communication system applicable to the embodiment of the present application;

Figure 2 is a schematic flow chart of the model effectiveness feedback method provided by the embodiment of the present application;

Figure 3 is another schematic flow chart of the model effectiveness feedback method provided by the embodiment of the present application;

Figure 4 is another schematic flow chart of the model effectiveness feedback method provided by the embodiment of the present application;

Figure 5 is another schematic flow chart of the model effectiveness feedback method provided by the embodiment of the present application;

Figure 6 is a schematic structural diagram of a model effectiveness feedback device provided by an embodiment of the present application;

Figure 7 is a schematic structural diagram of a communication device provided by an embodiment of the present application;

Figure 8 is a schematic structural diagram of a terminal that implements an embodiment of the present application;

Figure 9 is a schematic structural diagram of a network side device that implements an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be clearly described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of them. Example. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art fall within the scope of protection of this application.

The terms "first", "second", etc. in the description and claims of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances so that the embodiments of the present application can be practiced in sequences other than those illustrated or described herein, and that "first" and "second" are distinguished objects It is usually one type, and the number of objects is not limited. For example, the first object can be one or multiple. In addition, "and/or" in the description and claims indicates at least one of the connected objects, and the character "/" generally indicates that the related objects are in an "or" relationship.

It is worth pointing out that the technology described in the embodiments of this application is not limited to Long Term Evolution (LTE)/LTE Evolution (LTE-Advanced, LTE-A) systems, and can also be used in other wireless communication systems, such as code Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access, OFDMA), Single-carrier Frequency Division Multiple Access (SC-FDMA) and other systems. The terms "system" and "network" in the embodiments of this application are often used interchangeably, and the described technology can be used not only for the above-mentioned systems and radio technologies, but also for other systems and radio technologies. The following description describes a New Radio (NR) system for example purposes, and NR terminology is used in much of the following description, but these techniques can also be applied to applications other than NR system applications, such as 6th ^generation Generation, 6G) communication system. Figure 1 shows a block diagram of a wireless communication system to which embodiments of the present application are applicable. The wireless communication system includes a terminal 11 and a network side device 12. The terminal 11 may be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer), or a notebook computer, a personal digital assistant (Personal Digital Assistant, PDA), a palmtop computer, a netbook, or a super mobile personal computer. (ultra-mobile personal computer, UMPC), mobile Internet device (Mobile Internet Device, MID), augmented reality (augmented reality, AR)/virtual reality (VR) equipment, robots, wearable devices (Wearable Devices), vehicle user equipment (VUE), pedestrian terminals (Pedestrian User Equipment, PUE), smart homes (with wireless Home equipment with communication functions (such as refrigerators, TVs, washing machines or furniture, etc.), game consoles, personal computers (PCs), teller machines or self-service machines and other terminal-side devices, wearable devices include: smart watches, smart bracelets , smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. It should be noted that in the embodiments of this application The specific type of the terminal 11 is not limited. The network side device 12 may include an access network device or a core network device, where the access network device 12 may also be called a wireless access network device or a radio access network. RAN), wireless access network function or wireless access network unit. The access network device 12 may include a base station, a Wireless Local Area Network (Wireless Local Area Network, WLAN) access point or a WiFi node, etc. The base station may be called a Node B, Evolved Node B (eNB), access point, Base Transceiver Station (BTS), radio base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (Extended Service Set (ESS), home B-node, home evolved B-node, transmission and reception point (Transmission Reception Point, TRP) or some other suitable term in the field, as long as the same technical effect is achieved, the base station is not limited to Specific technical vocabulary, it should be noted that in the embodiment of this application, only the base station in the NR system is used as an example for introduction, and the specific type of the base station is not limited. The core network equipment may include but is not limited to at least one of the following: Core network Node, core network function, Mobility Management Entity (MME), Access and Mobility Management Function (AMF), Session Management Function (SMF), User Plane Function , UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Server Discovery Function (EASDF), Unified Data Management (Unified Data Management, UDM), Unified Data Repository (Unified Data Repository, UDR), Home Subscriber Server (HSS), Centralized network configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), local NEF (Local NEF, or L-NEF), binding support function (BSF), application function (AF), location management function (LMF), enhanced service mobile positioning center (Enhanced Serving Mobile Location Centre, E-SMLC), network data analytics function (NWDAF), etc. It should be noted that in the embodiment of this application, only the core network equipment in the NR system is used as an example for introduction, and the specific type of the core network equipment is not limited.

The model validity feedback, device, terminal and network-side equipment provided by the embodiments of the present application will be described in detail below with reference to the accompanying drawings through some embodiments and their application scenarios.

As shown in Figure 2, the embodiment of the present application provides a model validity feedback method. The execution subject of the method may be the first communication device. In other words, the method may be executed by software or hardware installed on the first communication device. . The method includes the following steps.

S210. The first communication device obtains first measurement information of N target terminals based on the first model, where N is a positive integer.

It should be understood that the first model can be a network architecture model obtained based on deep learning and/or machine exercises, such as a neural network (Neural Network). For the sake of simplicity, in the following embodiments, the model used for positioning is used. The AI model is taken as an example to illustrate.

It should be understood that the first communication device is a communication device capable of performing inference based on the first model, and may be a terminal or a network side device, such as a communication device having a location management function (Location Management Function, LMF).

It should be understood that the first measurement information of the N target terminals may be the first measurement information of one target terminal, or may be the first measurement information of multiple target terminals. Where N can be determined based on network configuration, protocol constraints, and algorithm implementation. It may also be determined based on the application scenario. For example, if the solution is used to determine the effectiveness of the first model of a single target terminal, N may be 1. If If the solution is used to determine the effectiveness of the first model for multiple target terminals, N may be greater than 1.

In one implementation, step S210 may include:

The first communication device obtains input information of the first model.

The input information of the first model can be obtained in a variety of ways, and the embodiments of this application only provide several specific implementation methods. In one implementation, the input information of the first model is collected through a data collection (Data Collection) process. The data collection process may be executed by the first communication device or may be executed by other communication devices and then sent to the first communication device.

In another embodiment, the input information of the first model is determined by a signal measurement process. The signal measurement process may be performed by the first communication device, or may be performed by other communication devices and then sent to the first communication device.

In another embodiment, the input information of the first model is determined through a channel estimation process. The channel estimation process may be performed by the first communication device, or may be performed by other communication devices and then sent to the first communication device.

For the sake of simplicity, in the following embodiments, the data collection process is used as an example to obtain the input information of the first model.

The input information of the first model may be diverse. In one implementation, the input information includes at least one of the following:

Signal measurement information for the target terminal;

The first location information of the target terminal, where the first location information is location information obtained through measurement;

error information;

Channel Impulse Response (CIR) information;

Delay power spectrum (Power Delay Profile, PDP) information.

In one implementation, the signal measurement information may include at least one of the following:

Reference Signal Time Difference (RSTD);

Round-Trip Time (RTT);

Angle of Arrival (AoA);

Angle of Departure (AoD);

Reference Signal Received Power (RSRP);

The power of the first path or multipath (path RSRP);

First path or multipath delay;

Time of Arrival (ToA) of the first path or multipath;

RSTD of first path or multipath;

Antenna subcarrier phase difference of first path or multipath;

Antenna subcarrier phase for first path or multipath;

Line Of Sight (LOS) indication information.

In one implementation, the first location information includes at least one of the following location information:

Absolute location information, such as latitude and longitude information;

Relative position information, such as the relative position to the measuring device or the Transmission and Reception Point (TRP).

In one implementation, the first location information is determined by at least one of the following:

Observed Time Difference of Arrival (OTDoA);

Global Navigation Satellite System (GNSS);

Downlink Time Difference of Arrival (DL-TDoA);

Uplink Time Difference of Arrival (UL-TDoA);

Uplink Angle of Arrival (UL-AoA);

departure angle;

Bluetooth;

sensor;

Wifi.

In one implementation, the error information includes at least one of the following:

position error value;

Measurement error value.

In one implementation, the channel impulse response information includes at least one of the following:

Channel impulse response in the time domain;

Channel impulse response in the frequency domain.

The channel impulse response information may also include processing information of the channel impulse response in the time domain or frequency domain, such as truncation information.

In one implementation, the channel impulse response information includes at least one of the following:

Channel impulse response of a single antenna;

Channel impulse response of multiple antennas;

Relative channel impulse response of multiple antennas.

After acquiring the input information of the first model, the first communication device may obtain first measurement information corresponding to the input information based on the first model based on the input information. Specifically, the acquired input information of the target terminal can be input into the first model, task inference can be performed based on the first model and the parameter information of the first model, and the first measurement information of the target terminal can be output. .

In one embodiment, if the input information of N target terminals is input respectively, then according to the A model determines first measurement information of the N target terminals.

The parameter information of the first model may include at least one of description information, hyperparameters and initial parameters of the first model.

According to the input information, the first measurement information obtained based on the first model may be diverse. In one implementation, the first measurement information may include at least one of the following:

RSTD;

ToA;

RTT;

AoA;

RSRP;

AoD.

In one embodiment, the first measurement information may also include LOS or NLOS identification information. Further LOS or NLOS information may be associated with first measurement information described below.

RSTD;

ToA;

RTT;

AoA;

RSRP;

AoD.

S220. The first communication device determines target information according to the first measurement information of the N target terminals; wherein the target information includes validity information of the relevant information of the first model.

The target information may include: model information of the first model, validity information of related information of the first model, or feedback information obtained after task inference based on the first model.

The first communication device may determine the validity of the first model according to the target information, and may also provide information to corresponding other communication devices, such as those having the model management (model Management) function of the communication device, sending corresponding feedback messages or request messages for updating the first model or requesting other models.

The feedback information obtained after performing task inference based on the first model may include performance evaluation information of the first model, the first communication device may determine the effectiveness of the first model based on the target information, and The performance evaluation information may also be sent to corresponding other communication devices, such as communication devices with the model management function, for updating the first model or requesting other models.

In one embodiment, the target information is determined according to the first measurement information associated with the LOS information of the N target terminals respectively.

In one embodiment, the target information is determined according to the first measurement information associated with the N target terminals and the NLOS information respectively.

It is worth noting that a variety of target information may be included to respectively indicate whether the first measurement information associated with the LOS information is valid or whether the first measurement information associated with the NLOS information is valid.

In one implementation, step S220 may include:

The target information is determined according to the first measurement information in R periods; wherein, R is a positive integer.

The first communication device may use the first measurement information obtained through multiple inferences to determine the target information, where the multiple inferences may be all inferences within a preset R period, or each inference corresponds to A cycle.

In another implementation manner, the first communication device may use the first measurement information obtained through multiple inferences, or multiple first measurement information obtained through multiple repetitions of inferences. Each inference Each corresponds to a repetition.

It can be known from the technical solutions of the above embodiments that in the embodiment of the present application, the first communication device obtains the first measurement information of the N target terminals based on the first model; the first communication device obtains the first measurement information of the N target terminals based on the first measurement information of the N target terminals. The information determines the target information, so that the effectiveness of the first model can be accurately obtained, and the first model can be monitored in real time to ensure that the first model can meet actual application requirements.

Based on the above embodiment, further, as shown in Figure 3, the S220 includes:

S221. Obtain second location information of the N target terminals based on the first measurement information of the N target terminals, where the second location information is location information obtained through calculation and/or estimation.

In one implementation, step S221 includes:

According to the first measurement information of the N target terminals by the M target TRPs, the second location information of the N target terminals is obtained, where M is a positive integer.

In an implementation manner, for each target terminal, the second location information of the target terminal can be obtained based on the plurality of first measurement information of the target terminal by the M target TRPs.

There can be various calculation methods for obtaining the second position information, and the embodiment of this application only provides an example of one of the specific implementation methods.

In one implementation, the first measurement information may include first time measurement information, such as ToA, RSTD, RTT, etc., and the second measurement information of the target terminal may be obtained based on time measurement information from multiple target TRPs to the target terminal. location information. For example, through DL-TDOA measurement, the following equations can be obtained:
R _i,1 =c(t _i -t ₁ )=R _i -R ₁

R _i is the distance from the i-th target TRP to the target terminal;

(x _i , y _i ) are the coordinates of the i-th target TRP, (x ₀ , y ₀ ) are the first position information of the target terminal;

N is the number of target TRPs, c is the speed of light, t _i -t ₁ is the arrival time difference between the i-th reference node and the first reference node;

The TDOA or RSTD of two target TRPs can locate the target terminal on a hyperbola;

The TDOA or RSTD of three or more target TRPs can determine the second location of the target terminal based on the position solution algorithm, such as least squares method, Taylor algorithm, Newton particle swarm, etc. location information.

In another embodiment, the first measurement information may include angle measurement information, and the second location information of the target terminal may be obtained based on angle measurement information from multiple target TRPs to the target terminal, such as AoA, AOD, etc. For example, when the target TRP is equipped with an antenna array, the antenna array can determine the AoA based on the signal sent by the target terminal. The AoA of the two TRPs are ∝ ₁ and ∝ ₂ respectively. Taking each target TRP as the starting point and AoA as the direction, a straight line is constructed intersection point to obtain the second location information of the target terminal. Assume that the position coordinates of the target terminal are (x, y) and the position coordinates of the M target TRPs are (x _i , y _i ). According to geometric meaning, the following formula is satisfied between them:

Expanding the above equation we get:
(xx _i )tan∝ _i =yy _i

y _i -x _i tan∝ _i =-xtan∝ _i +y

Y＝AX

The least square method can be used to solve X to obtain the second location information of the target terminal.

In another implementation manner, the second location information of the target terminal may be obtained based on time measurement information and angle measurement information from one or more target TRPs to the target terminal.

S222. Obtain second measurement information of the N target terminals according to the second location information of the N target terminals. The second measurement information is the measurement between the N target terminals and M target TRPs. Information, the M is a positive integer.

The second measurement information between the target terminal and the M target TRPs may be inferred based on the second location information of each target terminal and the location information of the M target TRPs respectively. Specifically, it may include at least one of the following:

According to the second location information of each target terminal and the location information of the M target TRPs, Infer the second time measurement information of the target terminal, such as RSTD, ToA, RTT, etc.;

According to the second location information of each target terminal and the location information of the M target TRPs, the second angle measurement information of the target terminal, such as AoA, AoD, etc., is inferred.

S223. Determine target information based on the first measurement information and second measurement information of the N target terminals.

Specifically, the target information may be determined based on the error information of the first measurement information and the second measurement information of the N target terminals.

In one implementation, step S223 includes:

The target information is determined based on the error information of the M target TRPs between the first measurement information and the second measurement information of the N target terminals.

The error information may include difference information between the first measurement information and the second measurement information of each target TRP for each target terminal.

In one implementation, the error information is error information of a single target terminal, wherein the error information of a single target terminal may include at least one of the following:

Difference information between the first measurement information and the second measurement information of one or more target TRPs for the target terminal;

At least one of the average error information, the maximum error information, or the median error information obtained based on the difference information between the first measurement information and the second measurement information obtained multiple times by the target terminal;

The distribution of error information obtained based on multiple error information of the target terminal, for example, based on the mean and variance of Gaussian distribution.

In another implementation, the error information may be error information of multiple target terminals, wherein the error information of the multiple target terminals may include at least one of the following:

Based on the error information of multiple target terminals, at least one of the average error information, the maximum error information, or the median error information is obtained;

The distribution of error information based on the error information of multiple target terminals, such as based on Gaussian score The mean and variance of the cloth.

In one implementation, the M target TRPs are the same as the target TRPs used to calculate the second location information of the target terminal in the above process.

The target TRPs can be selected in various ways. In one implementation, the M target TRPs are TRPs determined based on at least one of the following:

RSRP, for example, the RSRP measured by the target TRP on the target terminal is greater than or equal to the preset RSRP threshold;

Arrival time, for example, the arrival time measured by the target TRP to the target terminal is less than or equal to the preset arrival time threshold, or the M TRPs with the earliest arrival time;

The corresponding relationship with LOS or non-line of sight (Non Line Of Sight, NLOS), for example, the TRP that has the LOS corresponding relationship with the target terminal is used as the target TRP. Specifically, it can be based on the LOS identification information carried in the measurement information received from the TRP or NLOS identification information to determine;

Located within the angle measurement window;

Located within the time measurement window.

The value of M can be set according to actual needs, for example, it can be 1, 2, 3, 4, 5, 6, 8, 12, 16, 24, etc.

It should be understood that the M target TRPs may be TRPs corresponding to all the N target terminals; they may also be TRPs corresponding to each target terminal respectively, that is, each target terminal may correspond to different M targets respectively. The values of TRP and M can also be different. Each target terminal can select a different target TRP according to the above selection method. Each target terminal obtains the first measurement information based on the corresponding M target TRPs, and then obtains the second location information of the target terminal, infers the second measurement information of the target terminal and the corresponding M target TRPs, and then The first measurement information and the second measurement information of each target terminal are aggregated to determine the target information.

In one implementation, the target information can also be determined by summarizing the error information obtained multiple times. information, wherein the multiple times may correspond to multiple cycles or may be multiple repetitions. The step S223 includes:

The target information is determined according to the error information of the M target TRPs between the first measurement information and the second measurement information of the N target terminals in L periods; wherein, the L is a positive integer.

In one implementation, the target information determined by the first communication device includes validity information of the related information of the first model, wherein the validity information may be used to indicate at least one of the following:

Whether the relevant information is valid, for example, can be indicated by 1-bit indication information whether the relevant information is valid;

The degree of validity, for example, can use multiple bits to indicate a value between 0 and 1 to indicate the degree of validity;

The number of valid target TRPs. For example, the validity of the M target TRPs can be determined based on the error information of the M target TRPs to determine the number of valid target TRPs;

The number of invalid target TRPs, for example, the number of invalid target TRPs is determined according to the method described above;

The effective proportion can be determined, for example, based on the error information of N target terminals, or the proportion of valid times in the total number of inferences; or it can be based on the effective first measurement. The number of information is used to determine the validity of this inference result or to determine the proportion of valid times in the total number of inferences.

The proportion of invalidity, for example, determines the proportion of invalid times in the total number of inferences according to the method described above.

In another implementation, the target information determined by the first communication device may further include at least one of the following:

The model information of the first model, for example, includes the identification information of the first model;

Feedback.

The feedback information is fed back by the first model to other corresponding communication devices after completing inference on the N target terminals. In one implementation, the feedback information includes at least one of the following item:

The first measurement information;

The input information of the first model;

Second measurement information, the second measurement information is obtained based on the second location information of the N target terminals, and the second location information of the N target terminals is based on the first measurements of the N target terminals. information obtained;

Error information between the first measurement information and the second measurement information;

Average error information between the first measurement information and the second measurement information;

Median error information between the first measurement information and the second measurement information;

The maximum error information between the second measurement information and the second measurement information;

Update request information is used to request an update to the first model, or to request other models.

In one implementation, the feedback information includes model performance information of the first model, and the model performance information includes at least one of the following:

The estimation error of the first measurement information;

The variance of the first measurement information;

The maximum value of the first measurement information;

In one implementation, the update request information includes at least one of the following:

Model identification information;

Update request information for models and/or parameters that satisfy the first measurement information distribution;

Model type information;

Model parameter information.

In one implementation, before step S210, the method further includes:

The first communication device obtains model information of the first model.

In one implementation, the model information of the first model includes parameter information of the first model and a first condition corresponding to the parameter information, and the first condition is used to determine validity. The first communication device may use the first measurement information and the second measurement information of the N target terminals to The target information is determined based on the first condition based on the error information of the M target TRPs.

The first condition can be diverse. In one embodiment, the first condition is related to at least one of the following:

Error threshold;

Error variance threshold;

The threshold of the error mean;

Threshold for the number of TRPs;

Threshold for the number of target TRPs;

Failure conditions of the first model.

In one implementation, the parameter information of the first model further includes:

The application document of the first model;

The input data format of the first model, such as the input format of parameters and the input format of measurement information, etc.;

The output data format of the first model, such as the output format of parameters and the output format of measurement information, etc.;

Hyperparameter information of the first model, such as the activation model used, number of iterations, batch size, etc.;

The weight information of the first model, for example, the multiplicative coefficient and the additive coefficient of each neuron in the first model, that is, the weight and bias, etc.

In one implementation, the model information of the first model further includes at least one of the following:

The network structure information of the first model;

Network type information of the first model;

Optimizer information for the first model.

Wherein, the network structure information of the first model may include the network structure information used to construct the first model. Information related to the network structure. In one implementation, the network structure information of the first model includes at least one of the following:

Fully connected neural network;

convolutional neural network;

recurrent neural network;

residual network;

It includes the combination of multiple small networks, such as fully connected + convolution, convolution + residual, etc.;

The number of hidden layers;

How the input layer and hidden layer are connected;

How the hidden layers are connected;

The connection method between the hidden layer and the output layer;

The number of neurons in each layer;

The weight corresponding to each neuron is the multiplicative coefficient;

The bias corresponding to each neuron, that is, the additive coefficient;

The activation function corresponding to each neuron;

Complexity information.

The first model can be based on a fully connected neural network, a convolutional neural network, a recurrent neural network or a residual network, etc.; it can also be a combination of multiple small networks.

In one implementation, the network type information of the first model includes at least one of the following:

fully connected;

Based on the first communication device;

Assisted by said first communications device;

hybrid model;

Unsupervised models;

Supervision model.

The model information of the first model can be expressed in a variety of forms. In one embodiment, the model information of the first model can be expressed as a system parameter list. The model information of the first model includes at least the following: One item list:

Neuron information list of neural network;

A list of types and/or locations of activated neurons;

List of hyperparameter information;

List of loss function information.

Wherein, the neuron information includes at least one of the following:

neuron type;

The corresponding weight of the neuron;

The corresponding bias of the neuron.

It can be known from the technical solutions of the above embodiments that the embodiments of the present application obtain the second location information of the N target terminals based on the first measurement information of the N target terminals; and then obtain the second location information of the N target terminals based on the first measurement information of the N target terminals. The second location information is used to obtain the second measurement information of the N target terminals. The second measurement information is the measurement information between the N target terminals and the M target TRPs; according to the second measurement information of the N target terminals, The first measurement information and the second measurement information determine the target information, so that the effectiveness of the first model can be accurately obtained, and the first model can be monitored in real time.

Based on the above embodiments, further, as shown in Figure 4, after the first communication device determines the target information, the first communication device can perform subsequent operations based on the target information, such as performing operations on the first model. For updates or adjustments, the embodiments of this application only provide several implementation methods.

In one implementation, the method of the first model further includes:

The first communication device obtains update information of the first model for updating the first model.

The first communication device may send feedback information or update request information to the first model to a communication device with a model management function, so that the communication device retrains or adjusts the first model and transfers the first model to the communication device. Update information of a model is sent to the first communication device.

In one implementation, the parameters in the model information of the first model can be divided into variable parameters and fixed parameters. After the first communication device obtains the updated information of the first model, the method further includes :

The first communication device updates the variable parameters in the parameter information of the first model according to the update information.

In another implementation manner, the first communication device obtaining the model information of the first model may include: the first communication device obtaining K configurations of the model information of the first model.

In the case where the model information of the first model includes K configurations, the method further includes: the first communication device selects the model information of the first model from the K configurations according to the target information. target configuration.

In another implementation manner, the method further includes: the first communication device obtaining model information of the second model.

The first communication device may also request a second model from a communication device with a model management function based on the target information, or the communication device with a model management function may request a second model from the first communication device based on the received feedback information. The communication device replies with model information of the second model that can be used to perform inference.

It can be seen from the technical solutions of the above embodiments that the embodiment of the present application obtains the update of the first model information, thereby updating the first model in a timely manner according to the target information to better adapt to the current application environment and improve the effectiveness of the first model.

Based on the above embodiments, further, the communication systems used to implement the technical solutions of the above embodiments can be diverse. For example, the LMF and the model management function can be the same communication device or different communication devices, and the LFM and the model management function can be used to perform validity The first communication device determined may be the same communication device or a different communication device, and the embodiments of this application only provide several specific implementations.

In one implementation, taking the example that both model inference and validity judgment occur on the LMF, that is, the first communication device is the LMF, the model validity feedback method includes the following steps as shown in Figure 4.

A1. Obtain training data through the preset data collection process;

A2. Train the first model according to the training data;

A3. Send the model information of the trained first model to the first communication device for model deployment (Model Deployment);

A4. After starting the positioning task, obtain the input data of the first model through the preset data collection process;

A5. The first communication device performs model inference on the input data based on the first model;

A6. Obtain output data (output), which may include first measurement information.

A7. Based on the first measurement information, perform validity judgment and determine the target information;

A8. After completing model inference or completing validity judgment, feedback information can be sent to the model management module or data collection module;

A9. The model management module updates the first model or re-selects the model based on the feedback information;

A10. The first communication device obtains update information for updating the first model or obtaining the second model.

In another implementation, it is assumed that both model inference and validity judgment occur on the terminal, that is, the first communication device is a terminal, and the model validity feedback method is shown in Figure 5 .

B1. The network side device sends the model information of the first model to the target terminal;

B2. The target terminal performs model inference based on the first model and obtains the first measurement information;

B3. Obtain the second location information of the N target terminals and the second measurement information of the N target terminals based on the multiple first measurement information of the N target terminals;

B4. Based on the first measurement information and the second measurement information, make a validity judgment to obtain the target information;

B5. If it is determined to be invalid, send feedback information to the network side device;

B6. The network side device performs model update or model selection on the first model based on the feedback information, and based on the update result or selection result, the terminal sends model information again;

B7. If it is determined to be valid, continue to use the first model for model inference, and send the inference result and/or validity identifier to the network side device.

It can be seen from the technical solutions of the above embodiments that the embodiments of the present application execute model reasoning and validity judgment on different network-side devices or terminals, thereby enabling the model validity feedback method to be more flexibly deployed in the communication system.

For the model validity feedback method provided by the embodiments of the present application, the execution subject may be a model validity feedback device. In the embodiment of the present application, the model effectiveness feedback device performing the model effectiveness feedback method is used as an example to illustrate the model effectiveness feedback device provided by the embodiment of the present application.

As shown in Figure 6, the model validity feedback device includes: a model reasoning module 601 and a validity judgment module 602.

The model reasoning module 601 is used to obtain the first measurement information of the N target terminals based on the first model; the validity judgment module 602 is used for the first communication device to obtain the first measurement information of the N target terminals based on the first measurement information of the N target terminals. Determine target information; wherein the target information includes the first model Validity information of related information, the N is a positive integer.

Further, the model reasoning module 601 is used to:

Obtain input information of the first model;

According to the input information, first measurement information of N target terminals is obtained based on the first model.

Further, the input information includes at least one of the following:

Signal measurement information for the target terminal;

The first location information of the target terminal, where the first location information is location information obtained through measurement;

error information;

Channel impulse response information;

Delay power spectrum information.

Further, the signal measurement information includes at least one of the following:

RSTD;

RTT;

AoA;

AoD;

RSRP;

The power of the first path or multipath;

First path or multipath delay;

First path or multipath ToA;

RSTD of first path or multipath;

Antenna subcarrier phase difference of first path or multipath;

Antenna subcarrier phase for first path or multipath;

LOS indication information.

Further, the first location information includes at least one of the following location information:

Absolute location information;

Relative location information.

Further, the first location information is determined by at least one of the following:

Arrival time difference positioning method;

Global Navigation Satellite Systems;

Downstream arrival time difference;

Upward arrival time difference;

Upward arrival angle;

departure angle;

Bluetooth;

sensor;

Wifi.

Further, the validity judgment module 602 is also configured to determine the target information according to the first measurement information in R periods; wherein the R is a positive integer.

Further, the first measurement information includes at least one of the following:

RSTD;

TOA;

RTT;

AoA;

RSRP;

AoD.

Further, the first model is a network architecture model obtained based on deep learning and/or machine exercises.

It can be known from the technical solutions of the above embodiments that the embodiments of the present application obtain the first measurement information of N target terminals based on the first model; determine the target information based on the first measurement information of the N target terminals, thereby accurately obtaining all the The validity of the first model is described, and the first model is tested real-time monitoring to ensure that the first model can meet actual application requirements.

Based on the above embodiment, further, the validity judgment module 602 is used to:

Obtain second location information of the N target terminals according to the first measurement information of the N target terminals;

According to the second location information of the N target terminals, obtain second measurement information of the N target terminals, where the second measurement information is measurement information between the N target terminals and M target TRPs;

Determine target information according to the first measurement information and second measurement information of the N target terminals;

Wherein, the second position information is position information obtained through calculation and/or estimation, and M is a positive integer.

Further, the validity judgment module 602 is used to:

Obtain the second location information of the N target terminals based on the first measurement information of the M target TRPs on the N target terminals;

The first measurement information includes at least one of the following:

time measurement information;

Angle measurement information.

Further, the validity judgment module 602 is used to:

The target information is determined based on the error information of the M target TRPs between the first measurement information and the second measurement information of the N target terminals.

Further, the M target TRPs are TRPs determined based on at least one of the following:

RSRP;

Time of arrival;

Correspondence to LOS or NLOS;

Located within the angle measurement window;

Located within the time measurement window.

Further, the validity judgment module 602 is used to:

The target information is determined according to the error information of the M target TRPs between the first measurement information and the second measurement information of the N target terminals in L periods; wherein, the L is a positive integer.

Further, the validity information is used to indicate at least one of the following:

is it effective;

degree of effectiveness;

The number of valid target TRPs;

The number of invalid target TRPs;

Effective proportion;

Invalid proportion.

Further, the target information also includes at least one of the following:

Model information of the first model;

Feedback.

Further, the feedback information includes at least one of the following:

The first measurement information;

The input information of the first model;

Second measurement information, the second measurement information is obtained based on the second location information of the N target terminals, and the second location information of the N target terminals is based on the first measurements of the N target terminals. information obtained;

Error information between the first measurement information and the second measurement information;

Average error information between the first measurement information and the second measurement information;

Median error information between the first measurement information and the second measurement information;

The maximum error information between the second measurement information and the second measurement information;

Update request information is used to request an update to the first model, or to request other models.

Further, the update request information includes at least one of the following:

Model identification information;

Update request information for models and/or parameters that satisfy the first measurement information distribution;

Model type information;

Model parameter information.

Further, the error information includes at least one of the following:

position error value;

Measurement error value.

Further, the channel impulse response information includes at least one of the following:

Channel impulse response in the time domain;

Channel impulse response in the frequency domain.

Further, the channel impulse response information includes at least one of the following:

Channel impulse response of a single antenna;

Channel impulse response of multiple antennas;

Relative channel impulse response of multiple antennas.

Further, the model inference module 601 is also used to obtain model information of the first model.

Further, the model information of the first model includes parameter information of the first model and a first condition corresponding to the parameter information, and the first condition is used to determine validity.

Further, the first condition is related to at least one of:

Error threshold;

Error variance threshold;

The threshold of the error mean;

Threshold for the number of TRPs;

Threshold for the number of target TRPs;

Failure conditions of the first model.

Further, the parameter information of the first model also includes:

The application document of the first model;

The input data format of the first model;

The output data format of the first model;

Hyperparameter information of the first model;

The weight information of the first model.

Further, the model information of the first model also includes at least one of the following:

The network structure information of the first model;

Network type information of the first model;

Optimizer information for the first model.

Further, the network structure information of the first model includes at least one of the following:

Fully connected neural network;

convolutional neural network;

recurrent neural network;

residual network;

The combination of multiple small networks included;

The number of hidden layers;

How the input layer and hidden layer are connected;

How the hidden layers are connected;

The connection method between the hidden layer and the output layer;

The number of neurons in each layer;

The weight corresponding to each neuron;

the bias corresponding to each neuron;

The activation function corresponding to each neuron;

Complexity information.

Further, the network type information of the first model includes at least one of the following:

fully connected;

Based on the first communication device;

Assisted by said first communications device;

hybrid model;

Unsupervised models;

Supervision model.

Further, the model information of the first model includes at least one of the following lists:

Neuron information list of neural network;

A list of types and/or locations of activated neurons;

List of hyperparameter information;

List of loss function information.

Further, the neuron information includes at least one of the following:

neuron type;

The corresponding weight of the neuron;

The corresponding bias of the neuron.

It can be known from the technical solutions of the above embodiments that the embodiments of the present application obtain the second location information of the N target terminals based on the first measurement information of the N target terminals; and then obtain the second location information of the N target terminals based on the first measurement information of the N target terminals. The second location information is used to obtain the second measurement information of the N target terminals. The second measurement information is the measurement information between the N target terminals and the M target TRPs; according to the second measurement information of the N target terminals, The first measurement information and the second measurement information determine the target information, so that the effectiveness of the first model can be accurately obtained, and the first model can be monitored in real time.

Based on the above embodiment, further, the model inference module 601 is further configured to obtain update information of the first model.

Further, the model reasoning module 601 is also configured to update the variable parameters in the parameter information of the first model according to the update information.

Further, the model inference module 601 is used to obtain K configurations of model information of the first model.

Further, the model reasoning module 601 is also configured to select a target configuration of the model information of the first model from the K configurations according to the target information.

Further, the model reasoning module 601 is also used to obtain model information of the second model.

It can be seen from the technical solutions of the above embodiments that the embodiments of the present application obtain the update information of the first model, thereby updating the first model in a timely manner according to the target information, so as to better adapt to the current application environment and improve all aspects of the application. Describe the effectiveness of the first model.

The model validity feedback device in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or may be a component in the electronic device, such as an integrated circuit or chip. The electronic device may be a terminal or other devices other than the terminal. For example, terminals may include but are not limited to the types of terminals 11 listed above, and other devices may be servers, network attached storage (Network Attached Storage, NAS), etc., which are not specifically limited in the embodiment of this application.

The model validity feedback device provided by the embodiments of the present application can implement each process implemented by the method embodiments in Figures 2 to 5 and achieve the same technical effect. To avoid duplication, details will not be described here.

Optionally, as shown in Figure 7, this embodiment of the present application also provides a communication device 700, which includes a processor 701 and a memory 702. The memory 702 stores programs or instructions that can be run on the processor 701, for example. , when the communication device 700 is a terminal, when the program or instruction is executed by the processor 701, each step of the above model validity feedback method embodiment is implemented, and the same technical effect can be achieved. When the communication device 700 is a network-side device, when the program or instruction is executed by the processor 701, each step of the above-mentioned model validity feedback method embodiment is implemented, and the same technical effect can be achieved. To avoid duplication, the details are not repeated here.

An embodiment of the present application also provides a terminal, including a processor and a communication interface. The processor is configured to obtain first measurement information of N target terminals based on the first model, and determine target information based on the first measurement information of the N target terminals. , the communication interface is used to send feedback information. This terminal embodiment corresponds to the above-mentioned terminal-side method embodiment. Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this terminal embodiment, and can achieve the same technical effect. Specifically, FIG. 8 is a schematic diagram of the hardware structure of a terminal that implements an embodiment of the present application.

The terminal 800 includes but is not limited to: radio frequency unit 801, network module 802, audio output unit 803, input unit 804, sensor 805, display unit 806, user input unit 807, interface unit At least some components of the unit 808, the memory 809, the processor 810, and the like.

Those skilled in the art can understand that the terminal 800 may also include a power supply (such as a battery) that supplies power to various components. The power supply may be logically connected to the processor 810 through a power management system, thereby managing charging, discharging, and power consumption through the power management system. Management and other functions. The terminal structure shown in FIG. 8 does not constitute a limitation on the terminal. The terminal may include more or fewer components than shown in the figure, or some components may be combined or arranged differently, which will not be described again here.

It should be understood that in the embodiment of the present application, the input unit 804 may include a graphics processing unit (Graphics Processing Unit, GPU) 8041 and a microphone 8042. The graphics processor 8041 is responsible for the image capture device (GPU) in the video capture mode or the image capture mode. Process the image data of still pictures or videos obtained by cameras (such as cameras). The display unit 806 may include a display panel 8061, which may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like. The user input unit 807 includes a touch panel 8071 and at least one of other input devices 8072 . Touch panel 8071, also known as touch screen. The touch panel 8071 may include two parts: a touch detection device and a touch controller. Other input devices 8072 may include but are not limited to physical keyboards, function keys (such as volume control keys, switch keys, etc.), trackballs, mice, and joysticks, which will not be described again here.

In this embodiment of the present application, after receiving downlink data from the network side device, the radio frequency unit 801 can transmit it to the processor 810 for processing; in addition, the radio frequency unit 801 can send uplink data to the network side device. Generally, the radio frequency unit 801 includes, but is not limited to, an antenna, amplifier, transceiver, coupler, low noise amplifier, duplexer, etc.

Memory 809 may be used to store software programs or instructions as well as various data. The memory 809 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required for at least one function (such as a sound playback function, Image playback function, etc.) etc. Additionally, memory 809 may include volatile memory or non-volatile memory, or memory 809 may include both volatile and non-volatile memory. Among them, the non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electrically erasable programmable read-only memory (Electrically EPROM, EEPROM) or flash memory. Volatile memory can be random access memory (Random Access Memory, RAM), static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (Synch link DRAM) , SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DRRAM). Memory 809 in embodiments of the present application includes, but is not limited to, these and any other suitable types of memory.

The processor 810 may include one or more processing units; optionally, the processor 810 integrates an application processor and a modem processor, where the application processor mainly handles operations related to the operating system, user interface, application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the above modem processor may not be integrated into the processor 810.

Wherein, the processor 810 is configured to obtain the first measurement information of N target terminals based on the first model; the first communication device determines the target information according to the first measurement information of the N target terminals; wherein, the target The information includes validity information of the related information of the first model, and N is a positive integer.

Further, the radio frequency unit 801 is used to obtain the input information of the first model;

The processor 810 is configured to obtain first measurement information of N target terminals based on the first model according to the input information.

Further, the input information includes at least one of the following:

Signal measurement information for the target terminal;

The first location information of the target terminal, where the first location information is location information obtained through measurement;

error information;

Channel impulse response information;

Delay power spectrum information.

Further, the signal measurement information includes at least one of the following:

RSTD;

RTT;

AoA;

AoD;

RSRP;

The power of the first path or multipath;

First path or multipath delay;

First path or multipath ToA;

RSTD of first path or multipath;

Antenna subcarrier phase difference of first path or multipath;

Antenna subcarrier phase for first path or multipath;

LOS indication information.

Further, the first location information includes at least one of the following location information:

Absolute location information;

Relative location information.

Further, the first location information is determined by at least one of the following:

Arrival time difference positioning method;

Global Navigation Satellite Systems;

Downstream arrival time difference;

Upward arrival time difference;

Upward arrival angle;

departure angle;

Bluetooth;

sensor;

Wifi.

Further, the processor 810 is also configured to determine the target information according to the first measurement information in R periods; wherein the R is a positive integer.

Further, the first measurement information includes at least one of the following:

RSTD;

TOA;

RTT;

AoA;

RSRP;

AoD.

Further, the first model is a network architecture model obtained based on deep learning and/or machine exercises.

The embodiments of the present application can accurately obtain the effectiveness of the first model and monitor the first model in real time to ensure that the first model can meet actual application requirements.

In the above embodiment, further, the processor 810 is used to:

Obtain second location information of the N target terminals according to the first measurement information of the N target terminals;

According to the second location information of the N target terminals, obtain second measurement information of the N target terminals, where the second measurement information is measurement information between the N target terminals and M target TRPs;

Determine target information according to the first measurement information and second measurement information of the N target terminals;

Wherein, the second position information is position information obtained through calculation and/or estimation, and M is a positive integer.

Further, the processor 810 is used for:

According to the first measurement information of the M target TRPs on the N target terminals, the obtained The second location information of the N target terminals;

The first measurement information includes at least one of the following:

time measurement information;

Angle measurement information.

Further, the processor 810 is used for:

The target information is determined based on the error information of the M target TRPs between the first measurement information and the second measurement information of the N target terminals.

Further, the M target TRPs are TRPs determined based on at least one of the following:

RSRP;

Time of arrival;

Correspondence to LOS or NLOS;

Located within the angle measurement window;

Located within the time measurement window.

Further, the processor 810 is used for:

The target information is determined according to the error information of the M target TRPs between the first measurement information and the second measurement information of the N target terminals in L periods; wherein, the L is a positive integer.

Further, the validity information is used to indicate at least one of the following:

is it effective;

degree of effectiveness;

The number of valid target TRPs;

The number of invalid target TRPs;

Effective proportion;

Invalid proportion.

Further, the target information also includes at least one of the following:

Model information of the first model;

Feedback.

Further, the feedback information includes at least one of the following:

The first measurement information;

The input information of the first model;

Second measurement information, the second measurement information is obtained based on the second location information of the N target terminals, and the second location information of the N target terminals is based on the first measurements of the N target terminals. information obtained;

Error information between the first measurement information and the second measurement information;

Average error information between the first measurement information and the second measurement information;

Median error information between the first measurement information and the second measurement information;

The maximum error information between the second measurement information and the second measurement information;

Update request information is used to request an update to the first model, or to request other models.

Further, the update request information includes at least one of the following:

Model identification information;

Update request information for models and/or parameters that satisfy the first measurement information distribution;

Model type information;

Model parameter information.

Further, the error information includes at least one of the following:

position error value;

Measurement error value.

Further, the channel impulse response information includes at least one of the following:

Channel impulse response in the time domain;

Channel impulse response in the frequency domain.

Further, the channel impulse response information includes at least one of the following:

Channel impulse response of a single antenna;

Channel impulse response of multiple antennas;

Relative channel impulse response of multiple antennas.

Further, the radio frequency unit 801 is also used to obtain model information of the first model.

Further, the model information of the first model includes parameter information of the first model and a first condition corresponding to the parameter information, and the first condition is used to determine validity.

Further, the first condition is related to at least one of:

Error threshold;

Error variance threshold;

The threshold of the error mean;

Threshold for the number of TRPs;

Threshold for the number of target TRPs;

Failure conditions of the first model.

Further, the parameter information of the first model also includes:

The application document of the first model;

The input data format of the first model;

The output data format of the first model;

Hyperparameter information of the first model;

The weight information of the first model.

Further, the model information of the first model also includes at least one of the following:

The network structure information of the first model;

Network type information of the first model;

Optimizer information for the first model.

Further, the network structure information of the first model includes at least one of the following:

Fully connected neural network;

convolutional neural network;

recurrent neural network;

residual network;

The combination of multiple small networks included;

The number of hidden layers;

How the input layer and hidden layer are connected;

How the hidden layers are connected;

The connection method between the hidden layer and the output layer;

The number of neurons in each layer;

The weight corresponding to each neuron;

the bias corresponding to each neuron;

The activation function corresponding to each neuron;

Complexity information.

Further, the network type information of the first model includes at least one of the following:

fully connected;

Based on the first communication device;

Assisted by said first communications device;

hybrid model;

Unsupervised models;

Supervision model.

Further, the model information of the first model includes at least one of the following lists:

Neuron information list of neural network;

A list of types and/or locations of activated neurons;

List of hyperparameter information;

List of loss function information.

Further, the neuron information includes at least one of the following:

neuron type;

The corresponding weight of the neuron;

The corresponding bias of the neuron.

The embodiment of the present application can accurately obtain the effectiveness of the first model, and perform Perform real-time monitoring.

Based on the above embodiment, further, the radio frequency unit 801 is further configured to obtain update information of the first model.

Further, the processor 810 is further configured to update variable parameters in the parameter information of the first model according to the update information.

Further, the radio frequency unit 801 is configured to obtain K configurations of model information of the first model.

Further, the processor 810 is further configured to select a target configuration of the model information of the first model from the K configurations according to the target information.

Further, the radio frequency unit 801 is also used to obtain model information of the second model.

The embodiments of the present application can better adapt to the current application environment and improve the effectiveness of the first model.

Specifically, the embodiment of the present application also provides a network side device. As shown in Figure 9, the network side device 900 includes: a processor 901, a network interface 902, and a memory 903. The network interface 902 is, for example, a common public radio interface (CPRI).

Specifically, the network side device 900 in this embodiment of the present invention also includes: instructions or programs stored in the memory 903 and executable on the processor 901. The processor 901 calls the instructions or programs in the memory 903 to execute each of the steps shown in Figure 6. The method of module execution and achieving the same technical effect will not be described in detail here to avoid duplication.

Embodiments of the present application also provide a readable storage medium. Programs or instructions are stored on the readable storage medium. When the program or instructions are executed by a processor, each process of the above-mentioned model validity feedback method embodiment is implemented, and can To achieve the same technical effect, to avoid repetition, we will not repeat them here.

Wherein, the processor is the processor in the terminal described in the above embodiment. The readable storage medium includes computer readable storage media, such as computer read-only memory ROM, random access memory RAM, magnetic disk or optical disk, etc.

An embodiment of the present application further provides a chip, which includes a processor and a communication interface. The communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement each process of the above-mentioned model validity feedback method embodiment, and can achieve the same technical effect. To avoid repetition, the details will not be described here.

It should be understood that the chips mentioned in the embodiments of this application may also be called system-on-chip, system-on-a-chip, system-on-chip or system-on-chip, etc.

Embodiments of the present application further provide a computer program/program product. The computer program/program product is stored in a storage medium. The computer program/program product is executed by at least one processor to implement the above model validity feedback method. Each process of the embodiment can achieve the same technical effect, so to avoid repetition, it will not be described again here.

Embodiments of the present application also provide a model validity feedback system, including: a terminal and a network side device. The terminal can be used to perform the steps of the model validity feedback method as described above. The network side device can be used to perform the above steps. The steps of the model effectiveness feedback method.

It should be noted that, in this document, the terms "comprising", "comprises" or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or device that includes a series of elements not only includes those elements, It also includes other elements not expressly listed or inherent in the process, method, article or apparatus. Without further limitation, an element defined by the statement "comprises a..." does not exclude the presence of additional identical elements in a process, method, article or apparatus that includes that element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, but may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved. Functions may be performed, for example, the methods described may be performed in an order different from that described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better. implementation. Based on this understanding, this application The essence of the technical solution or the part that contributes to the existing technology can be embodied in the form of a computer software product. The computer software product is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes a number of instructions. It is used to cause a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in various embodiments of this application.

The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above-mentioned specific implementations. The above-mentioned specific implementations are only illustrative and not restrictive. Those of ordinary skill in the art will Inspired by this application, many forms can be made without departing from the purpose of this application and the scope protected by the claims, all of which fall within the protection of this application.

Claims (39)

A model effectiveness feedback method, including: The first communication device obtains first measurement information of N target terminals based on the first model; The first communication device determines target information based on the first measurement information of the N target terminals; Wherein, the target information includes validity information of the related information of the first model, and N is a positive integer. The method according to claim 1, wherein the determining target information according to the first measurement information of the N target terminals includes: Obtain second location information of the N target terminals according to the first measurement information of the N target terminals; According to the second location information of the N target terminals, second measurement information of the N target terminals is obtained, and the second measurement information is the measurement between the N target terminals and M target transmitting and receiving points. information; Determine target information according to the first measurement information and second measurement information of the N target terminals; Wherein, the second position information is position information obtained through calculation and/or estimation, and M is a positive integer. The method according to claim 2, wherein obtaining the second location information of the N target terminals based on the first measurement information of the N target terminals includes: Obtain the second location information of the N target terminals based on the first measurement information of the N target terminals from the M target transmitting and receiving points; The first measurement information includes at least one of the following: time measurement information; Angle measurement information. The method according to claim 2, wherein determining the target information according to the first measurement information and the second measurement information of the N target terminals includes: The target information is determined based on the error information of the M target transmitting and receiving points between the first measurement information and the second measurement information of the N target terminals. The method according to claim 2, wherein the M target transmission and reception points are transmission and reception points determined according to at least one of the following: Reference signal received power; Time of arrival; Correspondence with line-of-sight or non-line-of-sight; Located within the angle measurement window; Located within the time measurement window. The method according to claim 4, wherein determining the target information based on the error information of the M target transmitting and receiving points between the first measurement information and the second measurement information of the N target terminals includes: The target information is determined based on the error information of the M target transmitting and receiving points between the first measurement information and the second measurement information of the N target terminals in L periods; wherein, the L is a positive integer. The method of claim 1, wherein the validity information is used to indicate at least one of the following: is it effective; degree of effectiveness; The number of valid target sending and receiving points; Invalid number of target sending and receiving points; Effective proportion; Invalid proportion. The method according to claim 1, wherein the target information further includes at least one of the following: Model information of the first model; Feedback. The method according to claim 8, wherein the feedback information includes at least one of the following: The first measurement information; The input information of the first model; Second measurement information, the second measurement information is obtained based on the second location information of the N target terminals, and the second location information of the N target terminals is based on the first measurements of the N target terminals. information obtained; Error information between the first measurement information and the second measurement information; Average error information between the first measurement information and the second measurement information; Median error information between the first measurement information and the second measurement information; The maximum error information between the second measurement information and the second measurement information; Update request information is used to request an update to the first model, or to request other models. The method according to claim 9, wherein the update request information includes at least one of the following: Model identification information; Update request information for models and/or parameters that satisfy the first measurement information distribution; Model type information; Model parameter information. The method according to claim 1, wherein the first communication device obtains the first measurement information of N target terminals based on the first model, including: The first communication device obtains input information of the first model; The first communication device obtains first measurement information of N target terminals based on the first model according to the input information. The method according to claim 11, wherein the input information includes at least one of the following: Signal measurement information for the target terminal; The first location information of the target terminal, where the first location information is location information obtained through measurement; error information; Channel impulse response information; Delay power spectrum information. The method of claim 12, wherein the signal measurement information includes at least one of the following: Reference signal time difference; Round trip delay; angle of arrival; departure angle; Reference signal received power; The power of the first path or multipath; First path or multipath delay; Arrival time of first path or multipath; Reference signal time difference of first path or multipath; Antenna subcarrier phase difference of first path or multipath; Antenna subcarrier phase for first path or multipath; Sight distance indication information. The method of claim 12, wherein the first location information includes at least one of the following location information: Absolute location information; Relative location information. The method of claim 12, wherein the first location information is determined by at least one of the following: Arrival time difference positioning method; Global Navigation Satellite Systems; Downstream arrival time difference; Upward arrival time difference; Upward arrival angle; departure angle; Bluetooth; sensor; Wifi. The method of claim 14, wherein the error information includes at least one of the following: position error value; Measurement error value. The method according to claim 14, wherein the channel impulse response information includes at least one of the following: Channel impulse response in the time domain; Channel impulse response in the frequency domain. The method according to claim 14, wherein the channel impulse response information includes at least one of the following: Channel impulse response of a single antenna; Channel impulse response of multiple antennas; Relative channel impulse response of multiple antennas. The method according to claim 1, wherein before the first communication device obtains the first measurement information of N target terminals based on the first model, the method further includes: The first communication device obtains model information of the first model. The method of claim 19, wherein the model information of the first model includes parameter information of the first model and a first condition corresponding to the parameter information, the first condition being used to determine validity . The method of claim 19, wherein the first condition relates to at least one: Error threshold; Error variance threshold; The threshold of the error mean; Threshold for the number of sending and receiving points; The threshold for the number of target sending and receiving points; Failure conditions of the first model. The method according to claim 20, wherein the parameter information of the first model further includes: The application document of the first model; The input data format of the first model; The output data format of the first model; Hyperparameter information of the first model; The weight information of the first model. The method of claim 19, wherein the model information of the first model includes at least one of the following: The network structure information of the first model; Network type information of the first model; Optimizer information for the first model. The method according to claim 23, wherein the network structure information of the first model includes at least one of the following: Fully connected neural network; convolutional neural network; recurrent neural network; residual network; The combination of multiple small networks included; The number of hidden layers; How the input layer and hidden layer are connected; How the hidden layers are connected; The connection method between the hidden layer and the output layer; The number of neurons in each layer; The weight corresponding to each neuron; the bias corresponding to each neuron; The activation function corresponding to each neuron; Complexity information. The method of claim 23, wherein the network type information of the first model includes at least one of the following: fully connected; Based on the first communication device; Assisted by said first communications device; hybrid model; Unsupervised models; Supervision model. The method of claim 19, wherein the model information of the first model includes at least one of the following lists: Neuron information list of neural network; A list of types and/or locations of activated neurons; List of hyperparameter information; List of loss function information. The method of claim 26, wherein the neuron information includes at least one of the following: neuron type; The corresponding weight of the neuron; The corresponding bias of the neuron. The method according to claim 1, wherein after the first communication device determines the target information according to the first measurement information of the N target terminals, the method further includes: The first communication device obtains updated information of the first model. The method according to claim 28, wherein after the first communication device obtains the updated information of the first model, the method further includes: The first communication device updates the variable parameters in the parameter information of the first model according to the update information. The method according to claim 19, wherein the first communication device obtains the model information of the first model, including: The first communication device acquires K configurations of model information of the first model. The method according to claim 30, wherein after the first communication device determines the target information according to the first measurement information of the N target terminals, the method further includes: The first communication device selects a target configuration of the model information of the first model from the K configurations based on the target information. The method according to claim 28, wherein after the first communication device determines the target information according to the first measurement information of the N target terminals, the method further includes: The first communication device obtains model information of the second model. The method according to claim 1, wherein the determining target information according to the first measurement information of the N target terminals includes: The target information is determined according to the first measurement information in R periods; wherein, R is a positive integer. The method of claim 1, wherein the first measurement information includes at least one of the following: Reference signal time difference; Time of arrival; Round trip delay; angle of arrival; Reference signal received power; Departure angle. The method according to claim 1, wherein the first model is a network architecture model obtained based on deep learning and/or machine exercises. A model effectiveness feedback device, including: A model inference module, used to obtain the first measurement information of N target terminals based on the first model; A validity judgment module, configured for the first communication device to determine target information based on the first measurement information of the N target terminals; Wherein, the target information includes validity information of the related information of the first model, and N is a positive integer. A terminal, including a processor and a memory, the memory stores programs or instructions that can be run on the processor, and when the programs or instructions are executed by the processor, the implementation of any one of claims 1 to 35 is achieved. The steps of the model effectiveness feedback method described above. A network side device, including a processor and a memory. The memory stores programs or instructions that can be run on the processor. When the program or instructions are executed by the processor, any one of claims 1 to 35 is implemented. The steps of the model effectiveness feedback method described in the item. A readable storage medium on which programs or instructions are stored. When the programs or instructions are executed by a processor, the steps of the model validity feedback method according to any one of claims 1 to 35 are implemented.