Magnetic communication enhancement method and device based on motion posture and interference recognition

Abstract

The invention provides a magnetic communication enhancement method and device based on motion attitude and interference identification, and relates to the technical field of positioning. The method comprises the following steps: constructing a magnetic communication time sequence data set, and inputting the data set into a bidirectional data conversion module to obtain inverse signal query network and time sequence enhancement network input; obtaining inverse signal query network output characteristics according to the input and an inverse signal query network based on RepViT design; and obtaining antenna gain characteristics according to the input and inverse signal query network output characteristics and the time sequence enhancement network. According to the method, the magnetic communication enhancement network is trained through communication signals of the communication terminal with the transmitting array, the receiving array and the IMU under different postures and paths, the optimal gain of the receiving array and the transmitting array of the communication terminal under the current posture is obtained, the influence of the target motion posture and surrounding interference signals is further remarkably reduced, and the communication performance of the communication terminal is improved. And the gains of different transmitting-receiving antennas are adaptively adjusted based on the learning result, so that the communication quality is improved.

Description

Technical Field

[0001] The present invention relates to the field of positioning technology, and in particular to a method and device for enhancing magnetic communication based on motion posture and interference recognition. Background Art

[0002] Magnetic field communication is a communication method that uses magnetic field as a carrier for information transmission. It is currently widely used in scenarios such as smart Internet of Things, industrial automation, biomedicine, security monitoring, and energy management. Since magnetic field communication generally uses a lower communication frequency, it may be interfered by magnetic fields from electrical equipment, power lines, and geomagnetic factors. Shielding, reasonable circuit layout design, differential mode signal transmission, and digital signal processing are usually used to reduce the impact of interference signals. In addition, magnetic field communication is directional, and the relative position and direction between the transmitter and the receiver will affect the effect of communication. Therefore, it is necessary to take appropriate measures and technical means to solve problems such as magnetic field interference and directionality, and improve the reliability and stability of magnetic field communication. Summary of the invention

[0003] In order to solve the technical problems of magnetic field interference and directionality in the prior art, the embodiment of the present invention provides a method and device for enhancing magnetic communication based on motion posture and interference recognition. The technical solution is as follows:

[0004] On the one hand, a magnetic communication enhancement method based on motion posture and interference recognition is provided, the method is implemented by a magnetic communication enhancement device, and the method includes:

[0005] S1. Construct a magnetic communication time series dataset; wherein the magnetic communication time series dataset includes IMU outputs at different postures, receiving array channel quality parameters, and optimal array gain.

[0006] S2. Input the magnetic communication timing data set into the bidirectional digital conversion module to obtain the input of the inverse signal query network and the input of the timing enhancement network.

[0007] S3. Design an inverse signal query network based on the RepViT network, and obtain the output features of the inverse signal query network according to the input of the inverse signal query network and the inverse signal query network.

[0008] S4. According to the input of the timing enhancement network, the output characteristics of the inverse signal query network and the timing enhancement network, the antenna gain characteristics are obtained.

[0009] S5. Construct a magnetic communication enhancement network based on the bidirectional digital conversion module, the inverse signal query network and the timing enhancement network, and obtain the output of the magnetic communication enhancement network based on the output characteristics of the inverse signal query network, the antenna gain characteristics and the magnetic communication enhancement network.

[0010] S6. Train the magnetic communication enhancement network according to the magnetic communication time series data set to obtain a trained magnetic communication enhancement model.

[0011] S7. Obtain the posture data for magnetic communication enhancement, input the posture data into the trained magnetic communication enhancement model, and obtain the optimal array gain of the communication terminal receiving array and transmitting array under the current posture.

[0012] Optionally, constructing a magnetic communication time series data set in S1 includes:

[0013] The IMU output, receiving array channel quality parameters, and optimal array gain of the receiving array and transmitting array of the communication terminal are obtained when the communication terminal moves in different postures and different acquisition paths.

[0014] Among them, the communication terminal carries a transmitting array, a receiving array and an IMU.

[0015] Different motion sections in the acquisition path introduce various types of interference signals.

[0016] The data dimension of IMU output is 1*9; the data dimension of receiving array channel quality parameters is N*M, where N represents the number of sensors and M represents the number of channel parameters of each sensor; the data dimension of optimal array gain is N+K, where K represents the number of transmitting antennas.

[0017] Optionally, in S2, the magnetic communication time series data set is input into the bidirectional digital conversion module to obtain the input of the inverse signal query network and the input of the time series enhancement network, including:

[0018] Expand the receiving array channel quality parameters to obtain a 1*(N*M)-dimensional expanded vector, and concatenate the expanded vector with the IMU output to obtain a 1*(N*M+9)-dimensional input C of the timing enhancement network. mt ; Where N represents the number of sensors and M represents the number of channel parameters of each sensor.

[0019] The elements in the IMU output are multiplied element by element with the receiving array channel quality parameter to obtain 9 matrix channels. The 9 matrix channels are concatenated to obtain the input C of the N*M*9 dimensional inverse signal query network.

[0020] Optionally, the inverse signal query network is designed based on the RepViT network in S3, and the output features of the inverse signal query network are obtained according to the input of the inverse signal query network and the inverse signal query network, including:

[0021] The input C of the inverse signal query network is input into the convolution layer with a convolution kernel of 1*1 to obtain an N*M*3-dimensional output matrix; where N represents the number of sensors and M represents the number of channel parameters of each sensor.

[0022] The N*M*3 dimensional output matrix is ​​input into the RepViT network for magnetic signal feature extraction, and the inverse signal query network output feature CO with a dimension of 1*(N+K) is obtained; where K is the number of transmitting antennas.

[0023] Optionally, the temporal enhancement network includes a forget gate, an input gate, and a gain gate.

[0024] In S4, the antenna gain characteristics are obtained according to the input of the timing enhancement network, the output characteristics of the inverse signal query network and the timing enhancement network, including:

[0025] S41, the best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mt Splice in the transposed layer of the forget gate to obtain the spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector to obtain the multiplied feature matrix G M1 , for the feature matrix G M1 Perform convolution operation to obtain the convolution feature matrix G M2 , using the sigmoid activation function for the feature matrix G M2 Activate and obtain the feature matrix G after sigmoid activation M2 , the feature matrix G after sigmoid activation M2 With gain memory cells S t-1 Multiply them together to get the output gain memory cell S of the forget gate t1 .

[0026] S42, the best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mt Splice in the first branch of the input gate to obtain a spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector to obtain the multiplied feature matrix G M3 , for the feature matrix G M3 Perform convolution operation to obtain the convolution feature matrix G M4 , using the sigmoid activation function for the feature matrix G M4 Activate and obtain the feature matrix G after sigmoid activation M4 .

[0027] The best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mtSplice in the second branch of the input gate to obtain a spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector to obtain the multiplied feature matrix G M5 , for the feature matrix G M5 Perform convolution operation to obtain the convolution feature matrix G M6 , use Tanh activation function to transform the feature matrix G M6 Activate and obtain the feature matrix G after Tanh activation M6 .

[0028] The feature matrix G after sigmoid activation M4 And the feature matrix G after Tanh activation M6 Multiply the result of the multiplication with the output gain memory cell S of the forget gate t1 Perform element-by-element summation and pass the summation result through the fully connected layer to obtain the output gain memory cell S of the input gate t .

[0029] S43, the best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mt , input into the first branch of the gain gate for splicing, obtain the spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector, and obtain the multiplied feature matrix G M7 , for the feature matrix G M7 Perform convolution operation to obtain the convolution feature matrix G M8 , using the sigmoid activation function for the feature matrix G M8 Activate and obtain the feature matrix G after sigmoid activation M8 .

[0030] The output gain memory cell S of the input gate will be obtained t The inverse signal query network output feature CO is input into the second branch of the gain gate for splicing to obtain a spliced ​​one-dimensional vector, and the spliced ​​one-dimensional vector is multiplied by the transposed matrix of the spliced ​​one-dimensional vector to obtain the multiplied feature matrix CO 1 , for the feature matrix CO 1 Perform convolution operation to obtain the convolution feature matrix CO 2 , using Tanh activation function to transform the feature matrix CO 2 Activate and obtain the feature matrix CO after Tanh activation 2 .

[0031] The feature matrix G after sigmoid activation M8 And the feature matrix CO after Tanh activation2 Multiply them together and pass the multiplication result through the fully connected layer to obtain the optimal antenna gain feature G at the current time t t .

[0032] Optionally, obtaining the magnetic communication enhancement network output according to the inverse signal query network output characteristics, antenna gain characteristics and the magnetic communication enhancement network in S5 includes:

[0033] The inverse signal query network output feature CO and antenna gain feature G t Multiply them to get a feature matrix of (N+K)*(N+K) dimensions. The feature matrix is ​​processed by the maximum pooling layer and the fully connected layer to get the magnetic communication enhanced network output of 1*(N+K) dimension.

[0034] Among them, each element in the output of the magnetic communication enhancement network represents the optimal array gain CG of the receiving array and the transmitting array of the communication terminal in the current posture.

[0035] Optionally, the step of training the magnetic communication enhancement network according to the magnetic communication time series data set in S6 to obtain a trained magnetic communication enhancement model includes:

[0036] The inverse signal query network is trained according to the magnetic communication time series data set to obtain a trained inverse signal query network.

[0037] The internal parameters of the trained inverse signal query network are fixed, and the timing enhancement network is trained according to the magnetic communication timing data set to obtain the trained timing enhancement network.

[0038] The trained inverse signal query network and the trained timing enhancement network are used as pre-training models, and the magnetic communication enhancement network is trained according to the magnetic communication timing data set to obtain a trained magnetic communication enhancement model.

[0039] On the other hand, a magnetic communication enhancement device based on motion posture and interference recognition is provided, and the device is applied to a magnetic communication enhancement method based on motion posture and interference recognition, and the device includes:

[0040] The magnetic communication timing data set construction module is used to construct a magnetic communication timing data set; wherein the magnetic communication timing data set includes IMU outputs under different postures, receiving array channel quality parameters and optimal array gain.

[0041] The bidirectional digital conversion module is used to input the magnetic communication timing data set into the bidirectional digital conversion module to obtain the input of the inverse signal query network and the input of the timing enhancement network.

[0042] The inverse signal query network module is used to design an inverse signal query network based on the RepViT network, and obtain the output characteristics of the inverse signal query network according to the input of the inverse signal query network and the inverse signal query network.

[0043] The timing enhancement network module is used to obtain antenna gain characteristics according to the input of the timing enhancement network, the output characteristics of the inverse signal query network and the timing enhancement network.

[0044] The magnetic communication enhancement network module is used to construct a magnetic communication enhancement network based on the bidirectional digital conversion module, the inverse signal query network and the timing enhancement network, and obtain the magnetic communication enhancement network output based on the inverse signal query network output characteristics, antenna gain characteristics and the magnetic communication enhancement network.

[0045] The training module is used to train the magnetic communication enhancement network according to the magnetic communication time series data set to obtain a trained magnetic communication enhancement model.

[0046] The output module is used to obtain the posture data to be enhanced for magnetic communication, input the posture data into the trained magnetic communication enhancement model, and obtain the optimal array gain of the communication terminal receiving array and transmitting array under the current posture.

[0047] Optionally, the magnetic communication time series data set construction module is further used to:

[0048] The IMU output, receiving array channel quality parameters, and optimal array gain of the receiving array and transmitting array of the communication terminal are obtained when the communication terminal moves in different postures and different acquisition paths.

[0049] Among them, the communication terminal carries a transmitting array, a receiving array and an IMU.

[0050] Different motion sections in the acquisition path introduce various types of interference signals.

[0051] The data dimension of IMU output is 1*9; the data dimension of receiving array channel quality parameters is N*M, where N represents the number of sensors and M represents the number of channel parameters of each sensor; the data dimension of optimal array gain is N+K, where K represents the number of transmitting antennas.

[0052] Optionally, the bidirectional digital conversion module is further used for:

[0053] Expand the receiving array channel quality parameters to obtain a 1*(N*M)-dimensional expanded vector, and concatenate the expanded vector with the IMU output to obtain a 1*(N*M+9)-dimensional input C of the timing enhancement network. mt ; Where N represents the number of sensors and M represents the number of channel parameters of each sensor.

[0054] The elements in the IMU output are multiplied element by element with the receiving array channel quality parameter to obtain 9 matrix channels. The 9 matrix channels are concatenated to obtain the input C of the N*M*9 dimensional inverse signal query network.

[0055] Optionally, the reverse signal query network module is further used to:

[0056] The input C of the inverse signal query network is input into the convolution layer with a convolution kernel of 1*1 to obtain an N*M*3-dimensional output matrix; where N represents the number of sensors and M represents the number of channel parameters of each sensor.

[0057] The N*M*3 dimensional output matrix is ​​input into the RepViT network for magnetic signal feature extraction, and the inverse signal query network output feature CO with a dimension of 1*(N+K) is obtained; where K is the number of transmitting antennas.

[0058] Optionally, the temporal enhancement network includes a forget gate, an input gate, and a gain gate.

[0059] The timing enhancement network module is further used to:

[0060] S41, the best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mt Splice in the transposed layer of the forget gate to obtain the spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector to obtain the multiplied feature matrix G M1 , for the feature matrix G M1 Perform convolution operation to obtain the convolution feature matrix G M2 , using the sigmoid activation function for the feature matrix G M2 Activate and obtain the feature matrix G after sigmoid activation M2 , the feature matrix G after sigmoid activation M2 With gain memory cells S t-1 Multiply them together to get the output gain memory cell S of the forget gate t1 .

[0061] S42, the best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mt Splice in the first branch of the input gate to obtain a spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector to obtain the multiplied feature matrix G M3 , for the feature matrix G M3 Perform convolution operation to obtain the convolution feature matrix G M4, using the sigmoid activation function for the feature matrix G M4 Activate and obtain the feature matrix G after sigmoid activation M4 .

[0062] The best antenna gain G obtained at the previous time t-1 t-1 And the input C of the timing enhancement network at the current time t mt Splice in the second branch of the input gate to obtain a spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector to obtain the multiplied feature matrix G M5 , for the feature matrix G M5 Perform convolution operation to obtain the convolution feature matrix G M6 , use Tanh activation function to transform the feature matrix G M6 Activate and obtain the feature matrix G after Tanh activation M6 .

[0063] The feature matrix G after sigmoid activation M4 And the feature matrix G after Tanh activation M6 Multiply the result of the multiplication with the output gain memory cell S of the forget gate t1 Perform element-by-element summation and pass the summation result through the fully connected layer to obtain the output gain memory cell S of the input gate t .

[0064] S43, the best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mt , input into the first branch of the gain gate for splicing, obtain the spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector, and obtain the multiplied feature matrix G M7 , for the feature matrix G M7 Perform convolution operation to obtain the convolution feature matrix G M8 , using the sigmoid activation function for the feature matrix G M8 Activate and obtain the feature matrix G after sigmoid activation M8 .

[0065] The output gain memory cell S of the input gate will be obtained t The inverse signal query network output feature CO is input into the second branch of the gain gate for splicing to obtain a spliced ​​one-dimensional vector, and the spliced ​​one-dimensional vector is multiplied by the transposed matrix of the spliced ​​one-dimensional vector to obtain the multiplied feature matrix CO 1 , for the feature matrix CO 1Perform convolution operation to obtain the convolution feature matrix CO 2 , using Tanh activation function to transform the feature matrix CO 2 Activate and obtain the feature matrix CO after Tanh activation 2 .

[0066] The feature matrix G after sigmoid activation M8 And the feature matrix CO after Tanh activation 2 Multiply them together and pass the multiplication result through the fully connected layer to obtain the optimal antenna gain feature G at the current time t t .

[0067] Optionally, the magnetic communication enhanced network module is further used to:

[0068] The inverse signal query network output feature CO and antenna gain feature G t Multiply them to get a feature matrix of (N+K)*(N+K) dimensions. The feature matrix is ​​processed by the maximum pooling layer and the fully connected layer to get the magnetic communication enhanced network output of 1*(N+K) dimension.

[0069] Among them, each element in the output of the magnetic communication enhancement network represents the optimal array gain CG of the receiving array and the transmitting array of the communication terminal in the current posture.

[0070] Optionally, the training module is further configured to:

[0071] The inverse signal query network is trained according to the magnetic communication time series data set to obtain a trained inverse signal query network.

[0072] The internal parameters of the trained inverse signal query network are fixed, and the timing enhancement network is trained according to the magnetic communication timing data set to obtain the trained timing enhancement network.

[0073] The trained inverse signal query network and the trained timing enhancement network are used as pre-training models, and the magnetic communication enhancement network is trained according to the magnetic communication timing data set to obtain a trained magnetic communication enhancement model.

[0074] On the other hand, a magnetic communication enhancement device is provided, which includes: a processor; a memory, on which computer-readable instructions are stored, and when the computer-readable instructions are executed by the processor, any one of the above-mentioned magnetic communication enhancement methods based on motion posture and interference recognition is implemented.

[0075] On the other hand, a computer-readable storage medium is provided, in which at least one instruction is stored, and the at least one instruction is loaded and executed by a processor to implement any one of the above-mentioned magnetic communication enhancement methods based on motion posture and interference recognition.

[0076] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0077] In the present invention, the magnetic communication enhancement method based on motion posture and interference identification, whose core technology is the magnetic communication enhancement network, has a real-time and accurate calculation process compared with other traditional methods, can significantly reduce the influence of target motion posture and surrounding interference signals, and adaptively adjust the gains of different transmitting and receiving antennas based on learning results to improve communication quality.

[0078] The magnetic communication enhancement network is a dual-branch composite neural network. The inverse signal query network has a powerful feature fitting ability of the convolution structure. The timing enhancement network is suitable for capturing long-term dependencies in timing communication signals. In the timing enhancement network, convolution features are innovatively introduced to guide the convergence of timing features, realizing information interaction and deep fusion of the dual-branch network. Compared with a single network structure or an independent dual-branch structure, this structure can more comprehensively and accurately explore the magnetic sensing mechanism, thereby improving the evaluation accuracy of the optimal antenna gain. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0080] Figure 1 It is a flow chart of a magnetic communication enhancement method based on motion posture and interference recognition provided by an embodiment of the present invention;

[0081] Figure 2 is a network structure diagram of a forget gate provided by an embodiment of the present invention;

[0082] Figure 3 is a network structure diagram of an input gate provided by an embodiment of the present invention;

[0083] Figure 4 is a network structure diagram of a gain gate provided by an embodiment of the present invention;

[0084] Figure 5 is a network structure diagram of a magnetic communication enhancement model provided by an embodiment of the present invention;

[0085] Figure 6 is a flow chart of constructing a magnetic communication enhancement model provided by an embodiment of the present invention;

[0086] Figure 7 It is a block diagram of a magnetic communication enhancement device based on motion posture and interference recognition provided by an embodiment of the present invention;

[0087] Figure 8 It is a structural schematic diagram of a magnetic communication enhancement device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0088] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0089] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0090] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same. "of", "corresponding, relevant" and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference between them is not emphasized, the meanings they intend to express are the same.

[0091] In the embodiments of the present invention, sometimes the subscripts such as W 1 It may be written in non-subscript form such as W1. When the difference is not emphasized, the meaning is the same.

[0092] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0093] The embodiment of the present invention provides a magnetic communication enhancement method based on motion posture and interference recognition, which can be implemented by a magnetic communication enhancement device, which can be a terminal or a server. Figure 1 The flowchart of the magnetic communication enhancement method based on motion posture and interference recognition is shown in the figure. The processing flow of the method may include the following steps:

[0094] S1. Construct a magnetic communication time series dataset; wherein the magnetic communication time series dataset includes IMU outputs at different postures, receiving array channel quality parameters, and optimal array gain.

[0095] Optionally, the above step S1 may include:

[0096] The IMU output, receiving array channel quality parameters, and optimal array gain of the receiving array and transmitting array of the communication terminal are obtained when the communication terminal moves in different postures and different acquisition paths.

[0097] Among them, the communication terminal carries a transmitting array, a receiving array and an IMU.

[0098] Different motion sections in the acquisition path introduce various types of interference signals.

[0099] The data dimension of IMU output is 1*9; the data dimension of receiving array channel quality parameters is N*M, where N represents the number of sensors and M represents the number of channel parameters of each sensor; the data dimension of optimal array gain is N+K, where K represents the number of transmitting antennas.

[0100] In a feasible implementation, a magnetic communication data acquisition experiment is carried out to guide the communication terminal carrying the transmitting array, receiving array and IMU to move in different postures and acquisition paths, and introduce various types of interference signals in different movement sections. The real-time IMU (Inertial Measurement Unit) output (data dimension is 1×9, including speed, acceleration and magnetic field strength in each direction) and the channel quality parameters of the receiving array (data dimension is N×M, N represents the number of sensors, M is the number of channel parameters of each sensor, including signal strength, signal-to-noise ratio, bit error rate and other information) are recorded synchronously. At the same time, the optimal gain of the receiving array and the transmitting array of the communication terminal in the current posture is debugged and recorded (data dimension is N+K, K is the number of transmitting antennas). After obtaining a sufficient amount of communication data, the IMU output, receiving array channel quality parameters and optimal array gain in each posture are matched one by one to form a magnetic communication time series data set.

[0101] S2. Input the magnetic communication timing data set into the bidirectional digital conversion module to obtain the input of the inverse signal query network and the input of the timing enhancement network.

[0102] Optionally, the above step S2 may include:

[0103] Expand the receiving array channel quality parameters to obtain a 1*(N*M)-dimensional expanded vector, and concatenate the expanded vector with the IMU output to obtain a 1*(N*M+9)-dimensional input C of the timing enhancement network. mt ; Where N represents the number of sensors and M represents the number of channel parameters of each sensor.

[0104] The elements in the IMU output are multiplied element by element with the receiving array channel quality parameter to obtain 9 matrix channels. The 9 matrix channels are concatenated to obtain the input C of the N*M*9 dimensional inverse signal query network.

[0105] In a feasible implementation, the function of the bidirectional digital conversion module is to process the sensor data obtained at the communication terminal in real time into the input format of the inverse signal query network and the timing enhancement network. In the bidirectional digital conversion module, the output signal of the receiving array with a data dimension of N×M is expanded into a 1*(N*M)-dimensional vector and spliced ​​with the 1*9-dimensional IMU output to obtain a 1*(N*M+9)-dimensional vector Cmt as the input of the timing enhancement network; the 9 elements of the IMU output are respectively multiplied element by element with the receiving array signal matrix, and the resulting 9 matrix channels are spliced ​​to obtain an N*M*9-dimensional matrix C as the input of the inverse signal query network. The above is the internal operation of the bidirectional digital conversion module.

[0106] S3. Design an inverse signal query network based on the RepViT network, and obtain the output features of the inverse signal query network according to the input of the inverse signal query network and the inverse signal query network.

[0107] Optionally, the above step S3 may include:

[0108] The input C of the inverse signal query network is input into the convolution layer with a convolution kernel of 1*1 to obtain an N*M*3-dimensional output matrix; where N represents the number of sensors and M represents the number of channel parameters of each sensor.

[0109] The N*M*3 dimensional output matrix is ​​input into the RepViT network for magnetic signal feature extraction, and the inverse signal query network output feature CO with a dimension of 1*(N+K) is obtained; where K is the number of transmitting antennas.

[0110] In a feasible implementation, since the magnetic communication enhancement method needs to be deployed on the mobile terminal for data processing and has high requirements for data delay rate, a new lightweight backbone network RepViT is used as the inverse signal query main network. First, the N*M*9-dimensional data matrix C is input into the convolution layer with a convolution kernel of 1*1 to convert it into an N*M*3-dimensional matrix; then the matrix is ​​input into the RepViT network for magnetic signal feature extraction, where the output C in the RepViT network structure is 4 Set to (N+K), and obtain a one-dimensional vector CO with a dimension of 1*(N+K). The inverse signal query network relies on the strong fitting ability of layer-by-layer convolution to extract the effective information in the current moment sensor data into the one-dimensional vector CO.

[0111] S4. According to the input of the timing enhancement network, the output characteristics of the inverse signal query network and the timing enhancement network, the antenna gain characteristics are obtained.

[0112] Optionally, the timing enhancement network design mainly includes three parts: forget gate design, input gate design and gain gate design.

[0113] The above step S4 may include the following steps S41-S43:

[0114] S41, the best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mt Splice in the transposed layer of the forget gate to obtain the spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector to obtain the multiplied feature matrix G M1 , for the feature matrix G M1 Perform convolution operation to obtain the convolution feature matrix G M2 , using the sigmoid activation function for the feature matrix G M2 Activate and obtain the feature matrix G after sigmoid activation M2 , the feature matrix G after sigmoid activation M2 With gain memory cells S t-1 Multiply them together to get the output gain memory cell S of the forget gate t1 .

[0115] In a feasible implementation, Figure 2 As shown, the input of the forget gate is the optimal antenna gain G at the previous moment. t-1 , whose dimension is 1*(N+K), and the sensor data C output by the bidirectional digital conversion module at the current moment mt , whose dimension is 1*(N*M+9). In the transposition layer, the two one-dimensional inputs are concatenated to obtain a one-dimensional vector of 1*(N+K+N*M+9) dimensions; then the vector is multiplied with its own transposed matrix to obtain a (N+K+N*M+9)*(N+K+N*M+9)-dimensional feature matrix G M1 . M1 Perform convolution operation to achieve feature extraction and dimension adjustment, and obtain the (N+K)*(N+K)-dimensional feature matrix G M2 , and then use sigmoid to G M2 Activate. M2 The interference and redundant information in the input signal are abstracted by convolution, and the corresponding positions of the redundant information are set to 0. Then, the G M2 With gain memory cells S t-1 Multiply to eliminate the gain memory cell S t-1 The information that is irrelevant to the current gain adjustment task is obtained, and the (N+K)*(N+K)-dimensional feature S is obtained. t1 .

[0116] S42, the best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time tmt Splice in the first branch of the input gate to obtain a spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector to obtain the multiplied feature matrix G M3 , for the feature matrix G M3 Perform convolution operation to obtain the convolution feature matrix G M4 , using the sigmoid activation function for the feature matrix G M4 Activate and obtain the feature matrix G after sigmoid activation M4 .

[0117] The best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mt Splice in the second branch of the input gate to obtain a spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector to obtain the multiplied feature matrix G M5 , for the feature matrix G M5 Perform convolution operation to obtain the convolution feature matrix G M6 , use Tanh activation function to transform the feature matrix G M6 Activate and obtain the feature matrix G after Tanh activation M6 .

[0118] The feature matrix G after sigmoid activation M4 And the feature matrix G after Tanh activation M6 Multiply the result of the multiplication with the output gain memory cell S of the forget gate t1 Perform element-by-element summation and pass the summation result through the fully connected layer to obtain the output gain memory cell S of the input gate. t .

[0119] In a feasible implementation, the input of the input gate is the same as that of the forget gate, which is the optimal antenna gain G at the previous moment. t-1 , whose dimension is 1*(N+K), and the sensor data C output by the bidirectional digital conversion module at the current moment mt , whose dimension is 1*(N*M+9). The network structure of the input gate is as follows Figure 3 As shown, in the left branch, first set the antenna gain G t-1 With sensor data G t-1 Input the transposed layer, convolution layer and Sigmoid activation layer with a structure similar to the forget gate, and obtain the matrix G with dimensions of (N+K+N*M+9)*(N+K+N*M+9) in turn. M3 、(N+K)*(N+K)-dimensional feature matrix G M4 And the activated G M4 .

[0120] Similarly, in the right branch, the antenna gain G t-1 With sensor data C mt Input the transposed layer and convolution layer with similar structure to the left branch, and obtain the matrix G with dimension (N+K+N*M+9)*(N+K+N*M+9) in turn. M5 、(N+K)*(N+K)-dimensional feature matrix G M6 , and use Tanh activation function to G M6 Finally, the activated G M4 and G M6 Multiply, where after activation G M4 and G M6 They respectively represent whether the position is used for memory update and the specific input magnetic communication information corresponding to the position.

[0121] Finally, the product is multiplied by the output of the forget gate (N+K)*(N+K) dimensional gain memory cell S t1 Perform element-by-element summation and map it into a gain memory cell S with dimension 1*(N+K) through a fully connected layer t At this time, the gain memory cell S t An information update has been obtained, which contains both historical empirical magnetic communication gain information and current magnetic communication gain information.

[0122] S43, the best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mt , input into the first branch of the gain gate for splicing, obtain the spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector, and obtain the multiplied feature matrix G M7 , for the feature matrix G M7 Perform convolution operation to obtain the convolution feature matrix G M8 , using the sigmoid activation function for the feature matrix G M8 Activate and obtain the feature matrix G after sigmoid activation M8 .

[0123] The output gain memory cell S of the input gate will be obtained t The inverse signal query network output feature CO is input into the second branch of the gain gate for splicing to obtain a spliced ​​one-dimensional vector, and the spliced ​​one-dimensional vector is multiplied by the transposed matrix of the spliced ​​one-dimensional vector to obtain the multiplied feature matrix CO 1 , for the feature matrix CO 1 Perform convolution operation to obtain the convolution feature matrix CO 2 , using Tanh activation function to transform the feature matrix CO2 Activate and obtain the feature matrix CO after Tanh activation 2 .

[0124] The feature matrix G after sigmoid activation M8 And the feature matrix CO after Tanh activation 2 Multiply them together and pass the multiplication result through the fully connected layer to obtain the optimal antenna gain feature G at the current time t t .

[0125] In a feasible implementation, the gain gate has four inputs, namely: the best antenna gain G at the previous moment; t-1 , whose dimension is 1*(N+K); the sensor data C output by the bidirectional digital conversion module at the current moment mt , whose dimension is 1*(N*M+9); the gain memory cell S output by the input gate t , whose dimension is 1*(N+K); the inverse signal query network output feature CO, whose dimension is 1*(N+K). The network structure of the gain gate is as follows Figure 4 As shown, in the left branch, first set the antenna gain G t-1 With sensor data C mt Input the transposed layer, convolution layer and Sigmoid activation layer with a structure similar to the forget gate, and obtain the matrix G with dimensions of (N+K+N*M+9)*(N+K+N*M+9) in turn. M7 、(N+K)*(N+K)-dimensional feature matrix G M7 And the activated G M7 .

[0126] In the right branch, the gain memory cell S t The output feature CO of the inverse signal query network is input into the transposed layer and convolution layer with a similar structure to the left branch, and the matrix CO with a dimension of (N+K)*(N+K) is obtained in turn. 1 、(N+K)*(N+K)-dimensional feature matrix CO 2 , and the Tanh activation function is used to 2 Finally, the activated G M8 and CO 2 Multiply, where after activation CO 2 and G M8 Respectively represent the deep signal features extracted by the inverse signal query network and whether the corresponding deep signal is introduced. Finally, the product is mapped to the optimal antenna gain feature G at the current moment through the fully connected layer. t , the dimension is 1*(N+K). Compared with the traditional sequential neural network, the gain gate performs antenna gain feature G tWhen performing the algorithm, not only the characteristics of the time series communication data are taken into consideration, but also the deep signal features extracted by the inverse signal query network are introduced to effectively improve its ability to represent the gain signal at the current moment.

[0127] S5. Construct a magnetic communication enhancement network based on the bidirectional digital conversion module, the inverse signal query network and the timing enhancement network, and obtain the output of the magnetic communication enhancement network based on the output characteristics of the inverse signal query network, the antenna gain characteristics and the magnetic communication enhancement network.

[0128] Optionally, obtaining the magnetic communication enhancement network output according to the inverse signal query network output characteristics, antenna gain characteristics and the magnetic communication enhancement network in S5 includes:

[0129] The inverse signal query network output feature CO and antenna gain feature G t Multiply them to get a feature matrix of (N+K)*(N+K) dimensions. The feature matrix is ​​processed by the maximum pooling layer and the fully connected layer to get the magnetic communication enhanced network output of 1*(N+K) dimension.

[0130] Among them, each element in the output of the magnetic communication enhancement network represents the optimal array gain CG of the receiving array and the transmitting array of the communication terminal in the current posture.

[0131] In a feasible implementation, the overall network structure of the magnetic communication enhancement model is as follows: Figure 5 As shown, the construction process is as follows Figure 6 After the sensor data is output to the bidirectional digital conversion module, it is input into the inverse signal query network and the timing enhancement network in different data formats. The output of the inverse signal query network is CO, and the output of the timing enhancement network is the antenna gain characteristic G. t , the dimensions are all 1*(N+K). Among them, CO is also used as part of the input of the timing enhancement network to guide feature optimization. Next, the outputs of the two are multiplied to obtain a (N+K)*(N+K)-dimensional feature matrix, which is processed in sequence by the maximum pooling layer and the fully connected layer in the output head to obtain an output result CG with a dimension of 1*(N+K). Each element of the output result CG represents the optimal gain of the receiving array and the transmitting array of the communication terminal in the current posture. It is calculated based on the dual antenna gain prediction information of the inverse signal query network and the timing enhancement network, and has a more accurate prediction accuracy.

[0132] S6. Train the magnetic communication enhancement network according to the magnetic communication time series data set to obtain a trained magnetic communication enhancement model.

[0133] Optionally, the above step S6 may include:

[0134] The inverse signal query network is trained according to the magnetic communication time series data set to obtain a trained inverse signal query network.

[0135] The internal parameters of the trained inverse signal query network are fixed, and the timing enhancement network is trained according to the magnetic communication timing data set to obtain the trained timing enhancement network.

[0136] The trained inverse signal query network and the trained timing enhancement network are used as pre-training models, and the magnetic communication enhancement network is trained according to the magnetic communication timing data set to obtain a trained magnetic communication enhancement model.

[0137] In a feasible implementation, the above step S6 is the distributed generation of the magnetic communication enhancement model. First, the inverse signal query network is trained separately based on the magnetic communication time series data set to obtain the corresponding model; then, the internal parameters of the inverse signal query network are fixed, and the time series enhancement network is trained to obtain the corresponding model; finally, the parameter fixation is cancelled, and the results of the above two-step training models are used as pre-training models, and the magnetic communication enhancement network is trained as a whole based on the magnetic communication time series data set. After the training converges, the final magnetic communication enhancement model is obtained.

[0138] S7. Obtain the posture data for magnetic communication enhancement, input the posture data into the trained magnetic communication enhancement model, and obtain the optimal array gain of the communication terminal receiving array and transmitting array under the current posture.

[0139] In the embodiment of the present invention, the magnetic communication enhancement method based on motion posture and interference identification, whose core technology is the magnetic communication enhancement network, compared with other traditional methods, the calculation process is real-time and accurate, can significantly reduce the impact of target motion posture and surrounding interference signals, and adaptively adjust the gains of different transmitting and receiving antennas based on learning results to improve communication quality.

[0140] The magnetic communication enhancement network is a dual-branch composite neural network. The inverse signal query network has a powerful feature fitting ability of the convolution structure. The timing enhancement network is suitable for capturing long-term dependencies in timing communication signals. In the timing enhancement network, convolution features are innovatively introduced to guide the convergence of timing features, realizing information interaction and deep fusion of the dual-branch network. Compared with a single network structure or an independent dual-branch structure, this structure can more comprehensively and accurately explore the magnetic sensing mechanism, thereby improving the evaluation accuracy of the optimal antenna gain.

[0141] Figure 7 1 is a block diagram of a magnetic communication enhancement device based on motion gesture and interference recognition according to an exemplary embodiment, wherein the device is used in a magnetic communication enhancement method based on motion gesture and interference recognition. Figure 7 The device includes a magnetic communication time series data set construction module 310, a bidirectional digital conversion module 320, an inverse signal query network module 330, a time series enhancement network module 340, a magnetic communication enhancement network module 350, a training module 360 ​​and an output module 370. Among them:

[0142] The magnetic communication timing data set construction module 310 is used to construct a magnetic communication timing data set; wherein the magnetic communication timing data set includes IMU outputs at different postures, receiving array channel quality parameters and optimal array gain.

[0143] The bidirectional digital conversion module 320 is used to input the magnetic communication timing data set into the bidirectional digital conversion module to obtain the input of the inverse signal query network and the input of the timing enhancement network.

[0144] The inverse signal query network module 330 is used to design an inverse signal query network based on the RepViT network, and obtain the output characteristics of the inverse signal query network according to the input of the inverse signal query network and the inverse signal query network.

[0145] The timing enhancement network module 340 is used to obtain antenna gain characteristics according to the input of the timing enhancement network, the output characteristics of the inverse signal query network and the timing enhancement network.

[0146] The magnetic communication enhancement network module 350 is used to construct a magnetic communication enhancement network based on the bidirectional digital conversion module, the inverse signal query network and the timing enhancement network, and obtain the magnetic communication enhancement network output based on the inverse signal query network output characteristics, antenna gain characteristics and the magnetic communication enhancement network.

[0147] The training module 360 ​​is used to train the magnetic communication enhancement network according to the magnetic communication time series data set to obtain a trained magnetic communication enhancement model.

[0148] The output module 370 is used to obtain the posture data to be subjected to magnetic communication enhancement, input the posture data into the trained magnetic communication enhancement model, and obtain the optimal array gain of the communication terminal receiving array and transmitting array under the current posture.

[0149] In the embodiment of the present invention, the magnetic communication enhancement method based on motion posture and interference identification, whose core technology is the magnetic communication enhancement network, compared with other traditional methods, the calculation process is real-time and accurate, can significantly reduce the impact of target motion posture and surrounding interference signals, and adaptively adjust the gains of different transmitting and receiving antennas based on learning results to improve communication quality.

[0150] The magnetic communication enhancement network is a dual-branch composite neural network. The inverse signal query network has a powerful feature fitting ability of the convolution structure. The timing enhancement network is suitable for capturing long-term dependencies in timing communication signals. In the timing enhancement network, convolution features are innovatively introduced to guide the convergence of timing features, realizing information interaction and deep fusion of the dual-branch network. Compared with a single network structure or an independent dual-branch structure, this structure can more comprehensively and accurately explore the magnetic sensing mechanism, thereby improving the evaluation accuracy of the optimal antenna gain.

[0151] Figure 8 is a schematic diagram of the structure of a magnetic communication enhancement device provided by an embodiment of the present invention, such as Figure 8 As shown, the magnetic communication enhancement device may include the above Figure 7 The magnetic communication enhancement device based on motion gesture and interference recognition is shown. Optionally, the magnetic communication enhancement device 410 may include a first processor 2001.

[0152] Optionally, the magnetic communication enhancing device 410 may further include a memory 2002 and a transceiver 2003 .

[0153] The first processor 2001, the memory 2002 and the transceiver 2003 may be connected via a communication bus.

[0154] Combine the following Figure 8 The components of the magnetic communication enhancement device 410 are described in detail:

[0155] The first processor 2001 is the control center of the magnetic communication enhancement device 410, and may be a processor or a general term for multiple processing elements. For example, the first processor 2001 is one or more central processing units (CPUs), or an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention, such as one or more microprocessors (digital signal processors, DSPs), or one or more field programmable gate arrays (FPGAs).

[0156] Optionally, the first processor 2001 may execute various functions of the magnetic communication enhancement device 410 by running or executing a software program stored in the memory 2002 and calling data stored in the memory 2002 .

[0157] In a specific implementation, as an embodiment, the first processor 2001 may include one or more CPUs, such as Figure 8 CPU0 and CPU1 are shown in FIG.

[0158] In a specific implementation, as an embodiment, the magnetic communication enhancement device 410 may also include multiple processors, such as Figure 8The first processor 2001 and the second processor 2004 are shown in FIG. Each of these processors can be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here can refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0159] The memory 2002 is used to store the software program for executing the solution of the present invention, and the first processor 2001 controls the execution. The specific implementation method can refer to the above method embodiment, which will not be repeated here.

[0160] Optionally, the memory 2002 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 2002 may be integrated with the first processor 2001, or may exist independently and access the first processor 2001 through the interface circuit ( Figure 8 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0161] The transceiver 2003 is used to communicate with a network device or a terminal device.

[0162] Optionally, the transceiver 2003 may include a receiver and a transmitter ( Figure 8 The receiver is used to implement a receiving function, and the transmitter is used to implement a sending function.

[0163] Optionally, the transceiver 2003 may be integrated with the first processor 2001, or may exist independently and communicate with the first processor 2001 through the interface circuit ( Figure 8 (not shown) is coupled to the first processor 2001, which is not specifically limited in this embodiment of the present invention.

[0164] It should be noted that Figure 8 The structure of the magnetic communication enhancement device 410 shown in the figure does not constitute a limitation on the router, and the actual knowledge structure recognition device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0165] In addition, the technical effects of the magnetic communication enhancement device 410 can refer to the technical effects of the magnetic communication enhancement method based on motion posture and interference recognition in the above-mentioned method embodiment, which will not be repeated here.

[0166] It should be understood that the first processor 2001 in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0167] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0168] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When a computer instruction or computer program is loaded or executed on a computer, a process or function according to an embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. Computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0169] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0170] In the present invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0171] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0172] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0173] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0174] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0175] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0176] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0177] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, etc., various media that can store program codes.

[0178] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A magnetic communication enhancement method based on motion posture and interference recognition, characterized in that: The method comprises: S1. Constructing a magnetic communication time series data set; wherein the magnetic communication time series data set includes IMU output, receiving array channel quality parameters and optimal array gain under different postures; S2, inputting the magnetic communication time series data set into a bidirectional digital conversion module to obtain an input of an inverse signal query network and an input of a time series enhancement network; S3. Designing an inverse signal query network based on the RepViT network, and obtaining an output feature of the inverse signal query network according to the input of the inverse signal query network and the inverse signal query network; S4. Obtain antenna gain characteristics according to the input of the timing enhancement network, the output characteristics of the inverse signal query network, and the timing enhancement network; S5, constructing a magnetic communication enhancement network according to the bidirectional digital conversion module, the inverse signal query network and the timing enhancement network, and obtaining a magnetic communication enhancement network output according to the inverse signal query network output characteristics, the antenna gain characteristics and the magnetic communication enhancement network; S6. Training the magnetic communication enhancement network according to the magnetic communication time series data set to obtain a trained magnetic communication enhancement model; S7. Obtain posture data for magnetic communication enhancement, input the posture data into the trained magnetic communication enhancement model, and obtain the optimal array gain of the communication terminal receiving array and the transmitting array in the current posture.

2. The magnetic communication enhancement method based on motion posture and interference recognition according to claim 1 is characterized in that: The construction of the magnetic communication time series data set in S1 includes: Obtain the IMU output, receiving array channel quality parameters, and optimal array gain of the receiving array and transmitting array of the communication terminal when the communication terminal moves in different postures and different acquisition paths; Wherein, the communication terminal carries a transmitting array, a receiving array and an IMU; Different moving sections in the acquisition path introduce various types of interference signals; The data dimension of the IMU output is 1*9; the data dimension of the receiving array channel quality parameter is N*M, N represents the number of sensors, and M is the number of channel parameters of each sensor; the data dimension of the optimal array gain is N+K, and K is the number of transmitting antennas.

3. The magnetic communication enhancement method based on motion posture and interference recognition according to claim 1 is characterized in that: The step of inputting the magnetic communication time series data set into the bidirectional digital conversion module in S2 to obtain the input of the inverse signal query network and the input of the time series enhancement network includes: Expand the receiving array channel quality parameter to obtain a 1*(N*M)-dimensional expanded vector, and concatenate the expanded vector with the IMU output to obtain a 1*(N*M+9)-dimensional input C of the timing enhancement network mt ; Where N represents the number of sensors, and M is the number of channel parameters of each sensor; The elements in the IMU output are respectively multiplied element by element with the receiving array channel quality parameter to obtain 9 matrix channels, and the 9 matrix channels are spliced ​​to obtain an input C of an N*M*9-dimensional inverse signal query network.

4. The magnetic communication enhancement method based on motion posture and interference recognition according to claim 1 is characterized in that: The inverse signal query network is designed based on the RepViT network in S3, and the output features of the inverse signal query network are obtained according to the input of the inverse signal query network and the inverse signal query network, including: Inputting the input C of the inverse signal query network into a convolution layer with a convolution kernel of 1*1, obtaining an N*M*3-dimensional output matrix; wherein N represents the number of sensors, and M represents the number of channel parameters of each sensor; The N*M*3-dimensional output matrix is ​​input into the RepViT network to extract magnetic signal features, and an inverse signal query network output feature CO with a dimension of 1*(N+K) is obtained; wherein K is the number of transmitting antennas.

5. The magnetic communication enhancement method based on motion posture and interference recognition according to claim 1 is characterized in that: The temporal enhancement network includes a forget gate, an input gate and a gain gate; The step S4 of obtaining antenna gain characteristics according to the input of the timing enhancement network, the output characteristics of the inverse signal query network and the timing enhancement network includes: S41, the best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mt Splice in the transposed layer of the forget gate to obtain a spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector to obtain a multiplied feature matrix G M1 , for the feature matrix G M1 Perform convolution operation to obtain the convolution feature matrix G M2 , the sigmoid activation function is used to M2 Activate and obtain the feature matrix G after sigmoid activation M2 , the feature matrix G after sigmoid activation M2 With gain memory cell S t-1 Multiply them together to get the output gain memory cell S of the forget gate t1 ; S42, the best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mt Splice in the first branch of the input gate to obtain a spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector to obtain a multiplied feature matrix G M3 , for the feature matrix G M3 Perform convolution operation to obtain the convolution feature matrix G M4 , the sigmoid activation function is used to M4 Activate and obtain the feature matrix G after sigmoid activation M4 ; The best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mt Splice in the second branch of the input gate to obtain a spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector to obtain a multiplied feature matrix G M5 , for the feature matrix G M5 Perform convolution operation to obtain the convolution feature matrix G M6 , use Tanh activation function to the feature matrix G M6 Activate and obtain the feature matrix G after Tanh activation M6 ; The sigmoid activated feature matrix G M4 And the feature matrix G after Tanh activation M6 Multiply the result of the multiplication with the output gain memory cell S of the forget gate t1 Perform element-by-element summation and pass the summation result through the fully connected layer to obtain the output gain memory cell S of the input gate t ; S43, the best antenna gain G obtained at the last moment t-1 t-1 And the input C of the timing enhancement network at the current time t mt , input into the first branch of the gain gate for splicing, to obtain a spliced ​​one-dimensional vector, multiply the spliced ​​one-dimensional vector with the transposed matrix of the spliced ​​one-dimensional vector, to obtain the multiplied feature matrix G M7 , for the feature matrix G M7 Perform convolution operation to obtain the convolution feature matrix G M8 , the sigmoid activation function is used to M8 Activate and obtain the feature matrix G after sigmoid activation M8 ; The output gain memory cell S of the input gate will be obtained t and inputting the output feature CO of the inverse signal query network into the second branch of the gain gate for splicing to obtain a spliced ​​one-dimensional vector, multiplying the spliced ​​one-dimensional vector by the transposed matrix of the spliced ​​one-dimensional vector to obtain a multiplied feature matrix CO1, performing a convolution operation on the feature matrix CO1 to obtain a convolved feature matrix CO2, and using a Tanh activation function to activate the feature matrix CO2 to obtain a Tanh-activated feature matrix CO2; The sigmoid activated feature matrix G M8 Multiply it with the Tanh activated feature matrix CO2, and pass the multiplication result through the fully connected layer to obtain the optimal antenna gain feature G at the current time t t .

6. The magnetic communication enhancement method based on motion posture and interference recognition according to claim 1, characterized in that: The step of obtaining the magnetic communication enhancement network output according to the inverse signal query network output characteristic, the antenna gain characteristic and the magnetic communication enhancement network in S5 includes: The inverse signal query network output feature CO and the antenna gain feature G t Multiply them to obtain a feature matrix of (N+K)*(N+K) dimensions, and process the feature matrix through a maximum pooling layer and a fully connected layer to obtain a magnetic communication enhanced network output of 1*(N+K) dimension; Among them, each element in the output of the magnetic communication enhancement network represents the optimal array gain CG of the receiving array and the transmitting array of the communication terminal in the current posture.

7. The magnetic communication enhancement method based on motion posture and interference recognition according to claim 1 is characterized in that: The step of training the magnetic communication enhancement network according to the magnetic communication time series data set to obtain a trained magnetic communication enhancement model in S6 includes: Training the inverse signal query network according to the magnetic communication time series data set to obtain a trained inverse signal query network; Fixing the internal parameters of the trained inverse signal query network, training the timing enhancement network according to the magnetic communication timing data set, and obtaining a trained timing enhancement network; The trained inverse signal query network and the trained timing enhancement network are used as pre-trained models, and the magnetic communication enhancement network is trained according to the magnetic communication timing data set to obtain a trained magnetic communication enhancement model.

8. A magnetic communication enhancement device based on motion posture and interference recognition, the magnetic communication enhancement device based on motion posture and interference recognition is used to implement the magnetic communication enhancement method based on motion posture and interference recognition as claimed in any one of claims 1 to 7, characterized in that: The device comprises: A magnetic communication time series data set construction module, used to construct a magnetic communication time series data set; wherein the magnetic communication time series data set includes IMU outputs at different postures, receiving array channel quality parameters and optimal array gain; A bidirectional digital conversion module, used for inputting the magnetic communication time series data set into the bidirectional digital conversion module to obtain the input of the inverse signal query network and the input of the time series enhancement network; An inverse signal query network module, used to design an inverse signal query network based on the RepViT network, and obtain an inverse signal query network output feature according to the input of the inverse signal query network and the inverse signal query network; A timing enhancement network module, used to obtain antenna gain characteristics according to the input of the timing enhancement network, the output characteristics of the inverse signal query network and the timing enhancement network; A magnetic communication enhancement network module, used to construct a magnetic communication enhancement network according to the bidirectional digital conversion module, the inverse signal query network and the timing enhancement network, and obtain a magnetic communication enhancement network output according to the inverse signal query network output characteristics, the antenna gain characteristics and the magnetic communication enhancement network; A training module, used for training the magnetic communication enhancement network according to the magnetic communication time series data set to obtain a trained magnetic communication enhancement model; The output module is used to obtain the posture data to be enhanced for magnetic communication, input the posture data into the trained magnetic communication enhancement model, and obtain the optimal array gain of the communication terminal receiving array and transmitting array under the current posture.

9. A magnetic communication enhancement device, characterized in that: The magnetic communication enhancement device comprises: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, which can be called by a processor to execute the method according to any one of claims 1 to 7.

Images