A hypersurface positioning method and device based on nonlinear frame dictionary learning

Abstract

The present invention relates to a hypersurface positioning method and device based on nonlinear framework dictionary learning, aiming to solve the technical problem that the sparse representation and dictionary learning problems existing in the prior art are based on linear modeling, and the positioning problem cannot be fitted most efficiently by linear modeling, resulting in a decrease in positioning accuracy. The present invention includes: obtaining the positioning data of the training target in the target area with a label matrix; combining the label matrix method and the dictionary learning method to construct a dictionary learning model and a target classification model, and adding a nonlinear function based on the combination of the dictionary learning model and the target classification model to calculate the discriminant dictionary of the positioning data and the classification parameters in the target area; based on the discriminant dictionary, using a sparse coding method to obtain the sparse coefficients of the target to be located; based on the sparse coefficients, combined with the classification parameters, the position information of the target to be located is obtained.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless positioning, and in particular relates to a hypersurface positioning method and device based on nonlinear frame dictionary learning. Background Art

[0002] With the continuous improvement of science and technology, wireless positioning technology has become an indispensable emerging science and technology in our lives. For example, people can use the built-in wireless positioning technology in the map navigation APP to know their location, which is convenient for planning travel routes and finding destinations. In some special situations, such as preventing burglary, intruder detection, outdoor emergency search and rescue, target object trajectory tracking, etc., the target cannot be required to carry fixed or special equipment, and traditional wireless positioning technology cannot be applied, so the wireless positioning technology without carrying equipment came into being.

[0003] Since metasurface materials can convert "useless" noise or interference signals that are ubiquitous in space into "useful" signals, that is, when modeling the target area to be located, it can help users obtain more and more meaningful signal data. Based on this, the existing technology uses sparse coding algorithms as a wireless positioning technology for portable devices. Sparse coding is an important method in the field of artificial intelligence / machine learning. Its essence is to use an overcomplete "dictionary" and a sparse vector to efficiently represent data. The sparse coding algorithm can accurately classify samples by training the classifier and the dictionary together.

[0004] The prior art has the following technical problems:

[0005] 1. Sparse representation and dictionary learning problems are based on linear modeling, but linear modeling cannot be used to fit positioning problems most efficiently, resulting in a decrease in positioning accuracy. Summary of the invention

[0006] The present invention provides a hypersurface positioning method and device based on nonlinear framework dictionary learning, aiming to solve the technical problem that the sparse representation and dictionary learning problems existing in the above-mentioned prior art are based on linear modeling, and the positioning problem cannot be fitted most efficiently by linear modeling, resulting in a decrease in positioning accuracy.

[0007] The technical solution of the present invention to solve the above technical problems is as follows: a hypersurface positioning method based on nonlinear frame dictionary learning, comprising:

[0008] S1: Obtain the positioning data of the training target in the target area with the label matrix;

[0009] S2: constructing a dictionary learning model and a target classification model by combining the label matrix method and the dictionary learning method, and adding a nonlinear function based on the combination of the dictionary learning model and the target classification model to calculate the discriminant dictionary of the positioning data and the classification parameters in the target area;

[0010] S3: Based on the discriminant dictionary, a sparse coding method is used to obtain sparse coefficients of the target to be located in the positioning area;

[0011] S4: Based on the sparse coefficient and in combination with the classification parameter, the location information of the target to be located is obtained.

[0012] The beneficial effects of the present invention are as follows: the present invention first obtains the positioning data of the training target in the target area based on the signal transmitting device, signal receiving device and metasurface device arranged around the training target, and then realizes the analysis and processing of the signal by means of nonlinear framework dictionary learning, and can realize efficient fitting through nonlinear modeling, and finally obtains the precise positioning information of the target to be located.

[0013] Based on the above technical solution, the present invention can also be improved as follows.

[0014] Furthermore, the above-mentioned metasurface positioning method is implemented based on a metasurface device, a plurality of signal receiving devices and a plurality of signal transmitting devices; and before S1, it also includes:

[0015] S0: The signal receiving device, the signal transmitting device and the metasurface device are arranged around the target area, and the reflection angle of the metasurface device is adjusted by turning on each signal transmitting device in turn to obtain the reflection angle data of the metasurface device relative to each signal transmitting device.

[0016] The beneficial effect of adopting the above further scheme is that the present invention adjusts the reflection angle of the metasurface device so that the signal strength (RSS) received by the signal receiving device is maximized; finally, the optimal reflection angle of all signal transmitting devices with respect to the metasurface device is recorded. The reflection angle data is the angle at which the signal information captured by all signal receiving devices is the most (maximum signal strength (RSS), fastest arrival time (TOA), etc. can be selected according to requirements), and the angle at this time is recorded.

[0017] Furthermore, the above S0 is specifically:

[0018] S0.1: The signal receiving device, the signal transmitting device and the metasurface device are arranged around the target area, a signal transmitting device is controlled to be turned on, and the signal communication between the currently turned on signal transmitting device and all the signal receiving devices is modulated. After the modulation is completed, the currently turned on signal transmitting device is turned off, and the next signal transmitting device is controlled to be turned on, until each signal transmitting device completes the modulation of the signal communication with all the signal receiving devices;

[0019] S0.2: Control a signal transmitting device to turn on, and obtain the reflection angle data of the metasurface device relative to the currently turned on signal transmitting device by adjusting the reflection angle of the metasurface device;

[0020] S0.3: Control the currently turned-on signal transmitting device to turn off and control the next signal transmitting device to turn on, and repeat step S0.2 until the reflection angle data of the metasurface device relative to each signal transmitting device is obtained.

[0021] The beneficial effect of adopting the above further scheme is: the present application first divides the target area into rectangular grids with labels, and then determines the positions of the signal receiving device, the signal transmitting device and the metasurface device, so as to facilitate the acquisition of the positioning data of the training target when the training target moves in the target area during subsequent training.

[0022] Furthermore, the above S1 is specifically:

[0023] S1.1: Divide the target area into a plurality of rectangular grids of the same size, assign a label to each rectangular grid, obtain a label matrix for representing the position of the target area, and allow the training target to traverse each label in the label matrix;

[0024] S1.2: adjusting the transmittance of the metasurface device to a non-reflection threshold so that the metasurface device does not reflect the transmission signal transmitted by the signal transmitting device;

[0025] S1.3: Control a signal transmitting device to turn on, and use all signal receiving devices to obtain the transmission signal emitted by the currently turned on signal transmitting device when traversing each label in the label matrix based on the training target;

[0026] S1.4: Turn off the currently turned on signal transmitting device, control the next signal transmitting device to turn on, and repeat S1.3 until all signal receiving devices are used to obtain the transmission signal emitted by each signal transmitting device when traversing each label in the label matrix based on the training target;

[0027] S1.5: adjusting the transmittance of the metasurface device to a superposition threshold value, so that the metasurface device reflects the transmission signal emitted by the signal transmitting device to generate a reflection signal;

[0028] S1.6: Control a signal transmitting device to turn on, adjust the reflection angle of the metasurface device according to the corresponding reflection angle data based on the currently turned-on signal transmitting device, and use all signal receiving devices to obtain the superposition signal generated by the currently turned-on signal transmitting device and the metasurface device; wherein the superposition signal is the superposition of the transmission signal emitted by the currently turned-on signal transmitting device when traversing each label in the label matrix based on the training target and the reflection signal generated by the metasurface device;

[0029] S1.7: Turn off the currently turned-on signal transmitting device, control the next signal transmitting device to turn on, and repeat S1.6 until all signal receiving devices are used to obtain the superimposed signal generated by each signal transmitting device and the metasurface device;

[0030] S1.8: When the transmittance of the metasurface device is adjusted to a non-reflection threshold, the transmission signal and the superposition signal received by the signal receiving device are respectively subjected to complex Fourier transform to obtain a transmission spectrum and a superposition spectrum respectively, and the transmission spectrum is subtracted from the superposition spectrum to obtain a reflection spectrum of the reflection signal;

[0031] S1.9: Extract and save the peak values ​​of the first harmonic and the second harmonic in the emission spectrum and the reflection spectrum respectively, and save the labels corresponding to the training targets when the peak values ​​of the first harmonic and the second harmonic are extracted; the peak values ​​of the first harmonic, the second harmonic and the labels corresponding to the training targets are the positioning data of the training targets, and the peak values ​​of the first harmonic and the second harmonic are the signal data in the positioning data.

[0032] The beneficial effect of adopting the above further scheme is: when receiving a signal, the present application first adjusts the transmittance of the metasurface device to the maximum, at which time the metasurface device hardly reflects the signal, and the signal receiving device receives the signal that has not been reflected by the metasurface, and then adjusts the transmittance of the metasurface to the minimum, at which time the signal receiving device receives the superimposed signal, and then obtains the spectrum of the reflected signal through the complex Fourier transform method.

[0033] Furthermore, the above S2 is specifically:

[0034] S2.1: Construct a dictionary learning model and a target classification model, where:

[0035] The dictionary learning model is shown in the first formula (1):

[0036]

[0037] The target classification model is shown in the second formula (2):

[0038]

[0039] Specifically, is the signal data, represents a real matrix with K rows and M columns, represents the dictionary matrix, represents a real matrix with K rows and L columns, represents a sparse matrix, It represents a real number matrix with L rows and M columns, and subject to is the constraint condition. represents the i-th column of the sparse matrix, represents a real vector with L rows and 1 column, is the label matrix of the training target, represents a real matrix with N rows and M columns, represents the classification parameters of the classifier, represents a real matrix with N rows and L columns, T, K, N, M, L are hyperparameters, and K+N<L;

[0040] The above-mentioned first formula (1) and the second formula (2) are integrated to generate an integrated formula; wherein the integrated formula is shown in the third formula (3):

[0041]

[0042] Specifically, λ represents a hyperparameter greater than zero;

[0043] S2.2: Add a nonlinear function to the integration formula to generate a nonlinear integration formula; wherein the nonlinear integration formula is shown in the fourth formula (4):

[0044]

[0045] Specifically, f1 and f2 are both nonlinear functions;

[0046] S2.3: Find the inverse function of f1 and f2 in the nonlinear integration formula to generate an inverse function formula; wherein the inverse function formula is shown in the fifth formula (5):

[0047]

[0048] Specifically, f1 -1 and They are all inverse functions;

[0049] The fifth formula (5) is integrated to obtain the inverse function integration formula; wherein the inverse function integration formula is shown in the sixth formula (6):

[0050]

[0051] S2.4: Solve the inverse function integration formula using the dictionary learning method to obtain the initial discriminant dictionary Dic and the initial classification parameter Wt, and then calculate the discriminant dictionary of the positioning data and classification parameters within the target area in,

[0052] The discriminant dictionary is calculated as shown in the seventh formula (7);

[0053]

[0054] The classification parameters are calculated as shown in the eighth formula (8):

[0055]

[0056] Specifically, Represents the discriminant dictionary, dic1, dic2, ..., dic L are the columns in the initial discriminant dictionary Dic, ||dic1||2,||dic2||2,…,||dic L ||2 is the bi-norm of each column in the initial discriminant dictionary Dic, Represents classification parameters, wt1, wt2,…, wt L are the columns in the initial classification parameters Wt.

[0057] The beneficial effect of adopting the above further solution is that the present application solves the discriminant dictionary of the positioning data and the classification parameters in the target area by combining a nonlinear function, so that the position information of the target to be positioned can be obtained more accurately.

[0058] Further, in the above S3, the sparse coding method includes an OMP algorithm or a BP algorithm, and the sparse coefficient of the target to be located is calculated by using the OMP algorithm or the BP algorithm; wherein the sparse coefficient of the target to be located is calculated as shown in the ninth formula (9):

[0059]

[0060] Specifically, represents the sparse coefficient, f1 -1 represents the inverse function, Y target Indicates the target to be located. Represents the discriminant dictionary, sps is a sparse vector, and subject to is a constraint.

[0061] The beneficial effect of adopting the above further scheme is: the present application reduces the consumption of computing resources and storage units and improves the computing speed by using a sparse coding method.

[0062] Furthermore, in the above S4, the position information of the target to be located is calculated as shown in the tenth formula (10):

[0063]

[0064] Among them, loc target represents position information, f2 represents nonlinear function, represents the classification parameter, Represents the sparse coefficient.

[0065] The beneficial effect of adopting the above further solution is: the present application calculates the position information of the target to be located by combining the sparse coefficient with the classification parameter.

[0066] In a second aspect, in order to solve the above technical problems, the present invention also provides a hypersurface positioning device based on nonlinear frame dictionary learning, comprising:

[0067] Positioning data acquisition module: used to obtain the positioning data of the training target in the target area with the label matrix; and to obtain the signal data of the target to be located;

[0068] Calculation module: used to construct a dictionary learning model and a target classification model by combining the label matrix method and the dictionary learning method, and add a nonlinear function based on the combination of the dictionary learning model and the target classification model to calculate the discriminant dictionary of the positioning data and the classification parameters in the target area;

[0069] Sparse coefficient acquisition module: used to obtain the sparse coefficients of the target to be located by using a sparse coding method based on the discriminant dictionary;

[0070] Position information acquisition module: used to obtain the position information of the target to be located based on the sparse coefficient and the classification parameter.

[0071] In the third aspect, in order to solve the above-mentioned technical problems, the present invention also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the hypersurface positioning method based on nonlinear framework dictionary learning of the present application is implemented.

[0072] In a fourth aspect, in order to solve the above-mentioned technical problem, the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the hypersurface positioning method based on nonlinear frame dictionary learning of the present application is implemented.

[0073] Additional aspects and advantages of the present application will be partially given in the following description, which will become apparent from the following description, or will be understood through the practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1A schematic flow chart of a hypersurface positioning method based on nonlinear frame dictionary learning provided by an embodiment of the present invention;

[0075] Figure 2 A schematic diagram of the structure of a metasurface positioning device based on nonlinear frame dictionary learning provided by one embodiment of the present invention;

[0076] Figure 3 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention;

[0077] Figure 4 A diagram of a device deployment provided for an embodiment of the present invention;

[0078] Figure 5 A signal spectrum diagram provided for an embodiment of the present invention;

[0079] Figure 6 A noise waveform diagram provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0080] The principles and features of the present invention are described below. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0081] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0082] The embodiment of the present invention provides a possible implementation method, such as Figure 1 As shown, a flow chart of a method for super-surface positioning based on nonlinear frame dictionary learning is provided. The scheme can be executed by any electronic device, for example, a terminal device, or jointly executed by a terminal device and a server. For the convenience of description, the method provided by the embodiment of the present invention will be described below by taking the server as the execution subject as an example. Figure 1 The method may include the following steps:

[0083] S1: Obtain the positioning data of the training target in the target area with the label matrix;

[0084] S2: constructing a dictionary learning model and a target classification model by combining the label matrix method and the dictionary learning method, and adding a nonlinear function based on the combination of the dictionary learning model and the target classification model to calculate the discriminant dictionary of the positioning data and the classification parameters in the target area;

[0085] S3: Based on the discriminant dictionary, a sparse coding method is used to obtain sparse coefficients of the target to be located in the positioning area;

[0086] S4: Based on the sparse coefficient and in combination with the classification parameter, the location information of the target to be located is obtained.

[0087] Among them, the present invention first obtains the positioning data of the training target in the target area based on the signal transmitting device, signal receiving device and metasurface device set around the training target, and then realizes signal analysis and processing through nonlinear framework dictionary learning. Through nonlinear modeling, efficient fitting can be achieved, and finally the precise positioning information of the target to be located is obtained.

[0088] Optionally, the metasurface positioning method is implemented based on a metasurface device, multiple signal receiving devices, and multiple signal transmitting devices; and before S1, it also includes:

[0089] S0: The signal receiving device, the signal transmitting device and the metasurface device are arranged around the target area, and the reflection angle of the metasurface device is adjusted by turning on each signal transmitting device in turn to obtain the reflection angle data of the metasurface device relative to each signal transmitting device.

[0090] The present invention adjusts the reflection angle of the metasurface device so that the signal strength (RSS) received by the signal receiving device is maximized; finally, the optimal reflection angle of all signal transmitting devices with respect to the metasurface device is recorded. The reflection angle data is the angle at which the signal information captured by all signal receiving devices is the most (maximum signal strength (RSS), fastest arrival time (TOA), etc. can be selected according to requirements), and the angle at this time is recorded.

[0091] Optionally, S0 is:

[0092] S0 is specifically:

[0093] S0.1: The signal receiving device, the signal transmitting device and the metasurface device are arranged around the target area, a signal transmitting device is controlled to be turned on, and the signal communication between the currently turned on signal transmitting device and all the signal receiving devices is modulated. After the modulation is completed, the currently turned on signal transmitting device is turned off, and the next signal transmitting device is controlled to be turned on, until each signal transmitting device completes the modulation of the signal communication with all the signal receiving devices;

[0094] In this embodiment, 7 signal receiving devices, 8 signal transmitting devices and 1 intelligent metasurface are placed as follows: Figure 4 In the position shown.

[0095] S0.2: Control a signal transmitting device to turn on, and obtain the reflection angle data of the metasurface device relative to the currently turned on signal transmitting device by adjusting the reflection angle of the metasurface device;

[0096] S0.3: Control the currently turned-on signal transmitting device to turn off and control the next signal transmitting device to turn on, and repeat step S0.2 until the reflection angle data of the metasurface device relative to each signal transmitting device is obtained.

[0097] Among them, the present application first divides the target area into rectangular grids with labels, and then determines the positions of the signal receiving device, the signal transmitting device and the metasurface device, so as to facilitate the acquisition of the positioning data of the training target when the training target moves in the target area during subsequent training.

[0098] Optionally, S1 is:

[0099] S1 is specifically:

[0100] S1.1: Divide the target area into a plurality of rectangular grids of the same size, assign a label to each rectangular grid, obtain a label matrix for representing the position of the target area, and allow the training target to traverse each label in the label matrix;

[0101] like Figure 4 As shown, in this embodiment, the 7m×7m target positioning area is first divided into 36 rectangular grids of the same size, and each grid is labeled in turn (from 1 to 36).

[0102] S1.2: adjusting the transmittance of the metasurface device to a non-reflection threshold so that the metasurface device does not reflect the transmission signal transmitted by the signal transmitting device;

[0103] S1.3: Control a signal transmitting device to turn on, and use all signal receiving devices to obtain the transmission signal emitted by the currently turned on signal transmitting device when traversing each label in the label matrix based on the training target;

[0104] S1.4: Turn off the currently turned on signal transmitting device, control the next signal transmitting device to turn on, and repeat S1.3 until all signal receiving devices are used to obtain the transmission signal emitted by each signal transmitting device when traversing each label in the label matrix based on the training target;

[0105] S1.5: adjusting the transmittance of the metasurface device to a superposition threshold value, so that the metasurface device reflects the transmission signal emitted by the signal transmitting device to generate a reflection signal;

[0106] S1.6: Control a signal transmitting device to turn on, adjust the reflection angle of the metasurface device according to the corresponding reflection angle data based on the currently turned-on signal transmitting device, and use all signal receiving devices to obtain the superposition signal generated by the currently turned-on signal transmitting device and the metasurface device; wherein the superposition signal is the superposition of the transmission signal emitted by the currently turned-on signal transmitting device when traversing each label in the label matrix based on the training target and the reflection signal generated by the metasurface device;

[0107] S1.7: Turn off the currently turned-on signal transmitting device, control the next signal transmitting device to turn on, and repeat S1.6 until all signal receiving devices are used to obtain the superimposed signal generated by each signal transmitting device and the metasurface device;

[0108] S1.8: When the transmittance of the metasurface device is adjusted to a non-reflection threshold, the transmission signal and the superposition signal received by the signal receiving device are respectively subjected to complex Fourier transform to obtain a transmission spectrum and a superposition spectrum respectively, and the transmission spectrum is subtracted from the superposition spectrum to obtain a reflection spectrum of the reflection signal;

[0109] S1.9: Extract and save the peak values ​​of the first harmonic and the second harmonic in the emission spectrum and the reflection spectrum respectively, and save the labels corresponding to the training targets when the peak values ​​of the first harmonic and the second harmonic are extracted; the peak values ​​of the first harmonic, the second harmonic and the labels corresponding to the training targets are the positioning data of the training targets, and the peak values ​​of the first harmonic and the second harmonic are the signal data in the positioning data.

[0110] like Figure 5 As shown, in this embodiment, the peak values ​​of the first harmonic and the second harmonic in the signal spectrum diagram are used as the received signal strength (RSS), that is, signal data. In particular, affected by noise, the frequency domain waveforms of the first harmonic and the second harmonic sometimes have a second highest peak in addition to the highest peak, such as Figure 6 As shown; at this time, the peak is first compensated and then extracted as the received signal strength (RSS). The peak compensation formula used in this embodiment is c=b+δa, where a is the RSS data of the second peak, b is the RSS data of the peak, and δ=0.3;

[0111] Among them, when receiving a signal, the present application first adjusts the transmittance of the metasurface device to the maximum. At this time, the metasurface device hardly reflects the signal, and the signal receiving device receives the signal that has not been reflected by the metasurface. Then, the transmittance of the metasurface is adjusted to the minimum. At this time, the signal receiving device receives the superimposed signal, and then the spectrum of the reflected signal is obtained through the complex Fourier transform method.

[0112] Optionally, S2 is specifically:

[0113] S2.1: Construct a dictionary learning model and a target classification model, where:

[0114] The dictionary learning model is shown in the first formula (1):

[0115]

[0116] The target classification model is shown in the second formula (2):

[0117]

[0118] Specifically, is the signal data, represents a real matrix with K rows and M columns, represents the dictionary matrix, represents a real matrix with K rows and L columns, represents a sparse matrix, It represents a real number matrix with L rows and M columns, and subject to is the constraint condition. represents the i-th column of the sparse matrix, represents a real vector with L rows and 1 column, is the label matrix of the training target, represents a real matrix with N rows and M columns, represents the classification parameters of the classifier, represents a real matrix with N rows and L columns, T, K, N, M, L are hyperparameters, and K+N<L;

[0119] The above-mentioned first formula (1) and the second formula (2) are integrated to generate an integrated formula; wherein the integrated formula is shown in the third formula (3):

[0120]

[0121] Specifically, λ represents a hyperparameter greater than zero;

[0122] S2.2: Add a nonlinear function to the integration formula to generate a nonlinear integration formula; wherein the nonlinear integration formula is shown in the fourth formula (4):

[0123]

[0124] Specifically, f1 and f2 are both nonlinear functions;

[0125] In this embodiment, f1 and f2 are variants of the Tanh function, and the nonlinear integration formula becomes as shown in the eleventh formula (11), which is:

[0126]

[0127] Where f is a variant of the Tanh function;

[0128] S2.3: Find the inverse function of f in the nonlinear integration formula to generate an inverse function formula, which is shown in the fifth formula (5):

[0129]

[0130] Among them, f -1 is the inverse function;

[0131] The fifth formula (5) is integrated to obtain the inverse function integration formula, which is shown in the sixth formula (6):

[0132]

[0133] S2.4: The inverse function integration formula is solved by a dictionary learning method. In this embodiment, the K-SVD algorithm is used to obtain an initial discriminant dictionary Dic and an initial classification parameter Wt, and then the discriminant dictionary of the positioning data is calculated. and classification parameters within the target area The discriminant dictionary is calculated as shown in the seventh formula (7), and the classification parameter is calculated as shown in the eighth formula (8). The seventh formula (7) and the eighth formula (8) are:

[0134]

[0135]

[0136] in, Represents the discriminant dictionary, dic1, dic2, ..., dic L are the columns in the initial discriminant dictionary Dic, ||dic1||2,||dic2||2,…,||dic L ||2 is the bi-norm of each column in the initial discriminant dictionary Dic, Represents classification parameters, wt1, wt2,…, wt L are the columns in the initial classification parameters Wt.

[0137] Among them, the present application solves the discriminant dictionary of the positioning data and the classification parameters in the target area by combining a nonlinear function, so that the position information of the target to be positioned can be obtained more accurately.

[0138] Optionally, in S3, the sparse coding method includes an OMP algorithm or a BP algorithm. The OMP algorithm or the BP algorithm is used to calculate the sparse coefficient of the target to be located. The sparse coefficient of the target to be located is calculated as shown in the ninth formula (9), and the ninth formula (9) is:

[0139]

[0140] Specifically, represents the sparse coefficient, f -1 represents the inverse function, Y target Indicates the target to be located. Represents the discriminant dictionary, sps is a sparse vector, and subject to is a constraint.

[0141] Among them, the present application reduces the consumption of computing resources and storage units and improves the computing speed by using sparse coding methods.

[0142] Optionally, in S4, the position information of the target to be located is calculated as shown in the tenth formula (10), which is:

[0143]

[0144] Among them, loc target represents the position information, f represents the nonlinear function, represents the classification parameter, Represents the sparse coefficient.

[0145] In this embodiment, loc target The index corresponding to the largest element in is the label position of the target to be located. target =(0.1, 0.2, 0.7, 0.3, 0, ... 0), then the target to be located is located in the matrix grid with label 3.

[0146] Based on Figure 1 Based on the same principle as the method shown in , the embodiment of the present invention also provides a super surface positioning device based on nonlinear frame dictionary learning, such as Figure 2 As shown in , the super surface positioning device based on nonlinear frame dictionary learning may include:

[0147] Positioning data acquisition module: used to obtain the positioning data of the training target in the target area with the label matrix; and to obtain the signal data of the target to be located;

[0148] Calculation module: used to construct a dictionary learning model and a target classification model by combining the label matrix method and the dictionary learning method, and add a nonlinear function based on the combination of the dictionary learning model and the target classification model to calculate the discriminant dictionary of the positioning data and the classification parameters in the target area;

[0149] Sparse coefficient acquisition module: used to obtain the sparse coefficients of the target to be located by using a sparse coding method based on the discriminant dictionary;

[0150] Position information acquisition module: used to obtain the position information of the target to be located based on the sparse coefficient and the classification parameter.

[0151] Among them, the present application first obtains the positioning data of the training target with a label matrix, calculates the discriminant dictionary of the positioning data and the classification parameters in the target area, and then obtains the signal data of the target to be located, that is, obtains the peak value of the first harmonic and the peak value of the second harmonic of the target to be located, and calculates the label of the target to be located based on the signal data of the target to be located.

[0152] In this embodiment, the signal transmitting device is not only responsible for configuring the carrier frequency, IQ frequency, sampling frequency and type of transmitting waveform of the transmitting signal, but also needs to continuously and stably send wireless signals; the signal receiving device first needs to set the signal receiving mode according to the configuration of the signal transmitting device, such as the receiving frequency, the type of received information, etc., and secondly needs to be responsible for continuously and stably receiving the wireless signal, and finally needs to be responsible for transmitting the received signal to the memory and the processor, and after the reception is completed, feedback the end information. After receiving the end information, the memory and the processor will command the signal transmitting device to stop sending the signal; the metasurface device and its control device are responsible for adjusting the reflection angle and transmittance of the metasurface material when different transmitting devices transmit signals; the memory and the processor are responsible for receiving the signal and feedback information of the signal receiver, storing data and operation codes, etc.

[0153] The hypersurface positioning device based on nonlinear framework dictionary learning in the embodiment of the present invention can execute the hypersurface positioning method based on nonlinear framework dictionary learning provided in the embodiment of the present invention. The implementation principle is similar. The actions performed by each module and unit in the hypersurface positioning device based on nonlinear framework dictionary learning in each embodiment of the present invention correspond to the steps in the hypersurface positioning method based on nonlinear framework dictionary learning in each embodiment of the present invention. For the detailed functional description of each module of the hypersurface positioning device based on nonlinear framework dictionary learning, please refer to the description of the corresponding hypersurface positioning method based on nonlinear framework dictionary learning shown in the previous text, which will not be repeated here.

[0154] Among them, the above-mentioned hypersurface positioning device based on nonlinear framework dictionary learning can be a computer program (including program code) running in a computer device, for example, the hypersurface positioning device based on nonlinear framework dictionary learning is an application software; the device can be used to execute the corresponding steps in the method provided in an embodiment of the present invention.

[0155] In some embodiments, the metasurface positioning device based on nonlinear framework dictionary learning provided by an embodiment of the present invention can be implemented in a combination of software and hardware. As an example, the metasurface positioning device based on nonlinear framework dictionary learning provided by an embodiment of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the metasurface positioning method based on nonlinear framework dictionary learning provided by an embodiment of the present invention. For example, the processor in the form of a hardware decoding processor can adopt one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs) or other electronic components.

[0156] In other embodiments, the hypersurface positioning device based on nonlinear frame dictionary learning provided by the embodiments of the present invention can be implemented in software. Figure 2 A metasurface positioning device based on nonlinear framework dictionary learning stored in a memory is shown, which can be software in the form of programs and plug-ins, and includes a series of modules, including a positioning data acquisition module, a calculation module, a sparse coefficient acquisition module and a position information acquisition module, for implementing the method provided in an embodiment of the present invention.

[0157] The modules involved in the embodiments of the present invention may be implemented by software or hardware, wherein the name of a module does not limit the module itself in some cases.

[0158] Based on the same principle as the method shown in the embodiments of the present invention, an electronic device is also provided in the embodiments of the present invention, which may include but is not limited to: a processor and a memory; the memory is used to store a computer program; the processor is used to execute the method shown in any embodiment of the present invention by calling the computer program.

[0159] In an alternative embodiment, an electronic device is provided, such as Figure 3 As shown, Figure 3The electronic device shown includes: a processor and a memory. The processor and the memory are connected, such as through a bus. Optionally, the electronic device may also include a transceiver, which may be used for data exchange between the electronic device and other electronic devices, such as data transmission and / or data reception. It should be noted that in actual applications, the transceiver is not limited to one, and the structure of the electronic device does not constitute a limitation on the embodiments of the present invention.

[0160] The processor may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of the present invention. The processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0161] The bus may include a path to transmit information between the above components. The bus may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0162] The memory can be ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or it can be EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), 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 to this.

[0163] The memory is used to store the application code (computer program) for executing the solution of the present invention, and the execution is controlled by the processor. The processor is used to execute the application code stored in the memory to implement the content shown in the above method embodiment.

[0164] The electronic device may also be a terminal device. Figure 3 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0165] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding contents of the aforementioned method embodiment.

[0166] According to another aspect of the present invention, a computer program product or a computer program is also provided, the computer program product or the computer program comprising computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the methods provided in the implementation modes of the above-mentioned various embodiments.

[0167] Computer program code for performing the operations of the present invention may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0168] It should be understood that the flow charts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the module, the program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a sequence different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0169] The computer-readable storage medium provided by the embodiment of the present invention may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that may be used by or in conjunction with an instruction execution system, device, or device.

[0170] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device executes the method shown in the above embodiment.

[0171] The above description is only a preferred embodiment of the present invention and an explanation of the technical principles used. Those skilled in the art should understand that the disclosure scope involved in the present invention is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above disclosed concept. For example, the above features are replaced with the technical features with similar functions disclosed in the present invention (but not limited to) by each other.

Claims

1. A hypersurface positioning method based on nonlinear frame dictionary learning, wherein the hypersurface positioning method is implemented based on a hypersurface device, a plurality of signal receiving devices, and a plurality of signal transmitting devices, and is characterized in that: include: S0: placing the signal receiving device, the signal transmitting device and the metasurface device around the target area, adjusting the reflection angle of the metasurface device by turning on each signal transmitting device in turn, and obtaining reflection angle data of the metasurface device relative to each signal transmitting device; S1: Obtain the positioning data of the training target in the target area with the label matrix; S2: constructing a dictionary learning model and a target classification model by combining the label matrix method and the dictionary learning method, and adding a nonlinear function based on the combination of the dictionary learning model and the target classification model to calculate the discriminant dictionary of the positioning data and the classification parameters in the target area; S3: Based on the discriminant dictionary, a sparse coding method is used to obtain sparse coefficients of the target to be located in the positioning area; S4: Based on the sparse coefficient and in combination with the classification parameter, the location information of the target to be located is obtained.

2. A hypersurface positioning method based on nonlinear frame dictionary learning according to claim 1, characterized in that: S0 is specifically: S0.1: The signal receiving device, the signal transmitting device and the metasurface device are arranged around the target area, a signal transmitting device is controlled to be turned on, and the signal communication between the currently turned on signal transmitting device and all the signal receiving devices is modulated. After the modulation is completed, the currently turned on signal transmitting device is turned off, and the next signal transmitting device is controlled to be turned on, until each signal transmitting device completes the modulation of the signal communication with all the signal receiving devices; S0.2: Control a signal transmitting device to turn on, and obtain the reflection angle data of the metasurface device relative to the currently turned on signal transmitting device by adjusting the reflection angle of the metasurface device; S0.3: Control the currently turned-on signal transmitting device to turn off and control the next signal transmitting device to turn on, and repeat step S0.2 until the reflection angle data of the metasurface device relative to each signal transmitting device is obtained.

3. A hypersurface positioning method based on nonlinear frame dictionary learning according to claim 2, characterized in that: S1 is specifically: S1.1: Divide the target area into a plurality of rectangular grids of the same size, assign a label to each rectangular grid, obtain a label matrix for representing the position of the target area, and allow the training target to traverse each label in the label matrix; S1.2: adjusting the transmittance of the metasurface device to a non-reflection threshold so that the metasurface device does not reflect the transmission signal transmitted by the signal transmitting device; S1.3: Control a signal transmitting device to turn on, and use all signal receiving devices to obtain the transmission signal emitted by the currently turned on signal transmitting device when traversing each label in the label matrix based on the training target; S1.4: Turn off the currently turned on signal transmitting device, control the next signal transmitting device to turn on, and repeat S1.3 until all signal receiving devices are used to obtain the transmission signal emitted by each signal transmitting device when traversing each label in the label matrix based on the training target; S1.5: adjusting the transmittance of the metasurface device to a superposition threshold value, so that the metasurface device reflects the transmission signal emitted by the signal transmitting device to generate a reflection signal; S1.6: Control a signal transmitting device to turn on, adjust the reflection angle of the metasurface device according to the corresponding reflection angle data based on the currently turned-on signal transmitting device, and use all signal receiving devices to obtain the superposition signal generated by the currently turned-on signal transmitting device and the metasurface device; wherein the superposition signal is the superposition of the transmission signal emitted by the currently turned-on signal transmitting device when traversing each label in the label matrix based on the training target and the reflection signal generated by the metasurface device; S1.7: Turn off the currently turned-on signal transmitting device, control the next signal transmitting device to turn on, and repeat S1.6 until all signal receiving devices are used to obtain the superimposed signal generated by each signal transmitting device and the metasurface device; S1.8: When the transmittance of the metasurface device is adjusted to a non-reflection threshold, a complex Fourier transform is performed on the transmission signal obtained by the signal receiving device to obtain a transmission spectrum, a complex Fourier transform is performed on the superimposed signal to obtain a superimposed spectrum, and the transmission spectrum is subtracted from the superimposed spectrum to obtain a reflection spectrum of the reflected signal; S1.9: Extract and save the peak values ​​of the first harmonic and the second harmonic in the emission spectrum and the reflection spectrum respectively, and save the labels corresponding to the training targets when the peak values ​​of the first harmonic and the second harmonic are extracted; the peak values ​​of the first harmonic, the second harmonic and the labels corresponding to the training targets are the positioning data of the training targets, and the peak values ​​of the first harmonic and the second harmonic are the signal data in the positioning data.

4. The method for super-surface localization based on nonlinear frame dictionary learning according to claim 3, characterized in that: S2 is specifically: S2.1: Construct a dictionary learning model and a target classification model, where: The dictionary learning model is shown in the first formula (1): The target classification model is shown in the second formula (2): Specifically, is the signal data, represents a real matrix with K rows and M columns, represents the dictionary matrix, represents a real matrix with K rows and L columns, represents a sparse matrix, represents a real number matrix with L rows and M columns, subject to is the constraint condition, represents the i-th column of the sparse matrix, represents a real vector with L rows and 1 column, is the label matrix of the training target, represents a real matrix with N rows and M columns, represents the classification parameters of the classifier, represents a real matrix with N rows and L columns. T, K, N, M, and L are hyperparameters and satisfy K+N <L; The first formula (1) and the second formula (2) are integrated to generate an integrated formula; wherein the integrated formula is shown in the third formula (3): Specifically, λ represents a hyperparameter greater than zero; S2.2: Add a nonlinear function to the integration formula to generate a nonlinear integration formula; wherein the nonlinear integration formula is shown in the fourth formula (4): Specifically, f1 and f2 are both nonlinear functions; S2.3: Find the inverse function of f1 and f2 in the nonlinear integration formula to generate an inverse function formula; wherein the inverse function formula is shown in the fifth formula (5): Specifically, and They are all inverse functions; The fifth formula (5) is integrated to obtain the inverse function integration formula; wherein the inverse function integration formula is shown in the sixth formula (6): S2.4: Solve the inverse function integration formula using the dictionary learning method to obtain the initial discriminant dictionary Dic and the initial classification parameter Wt, and then calculate the discriminant dictionary of the positioning data and classification parameters within the target area in, Calculate the discriminant dictionary As shown in the seventh formula (7): Calculate the classification parameters As shown in the eighth formula (8): Specifically, Represents the discriminant dictionary, dic1, dic2, ..., dic L are the columns in the initial discriminant dictionary Dic, ∥dic1∥2,∥dic2∥2,…,∥dic L ∥2 is the bi-norm of each column in the initial discriminant dictionary Dic, Represents classification parameters, wt1, wt2,…, wt L are the columns in the initial classification parameters Wt.

5. The method for super-surface localization based on nonlinear frame dictionary learning according to claim 1, characterized in that: In S3, the sparse coding method includes an OMP algorithm or a BP algorithm, and the sparse coefficient of the target to be located is calculated by using the OMP algorithm or the BP algorithm; wherein the sparse coefficient of the target to be located is calculated as shown in the ninth formula (9): Specifically, represents the sparse coefficient, represents the inverse function, Y target Indicates the target to be located. Represents the discriminant dictionary, sps is a sparse vector, and subject to is a constraint.

6. The method for super-surface localization based on nonlinear frame dictionary learning according to claim 1, characterized in that: In S4, the position information of the target to be located is calculated as shown in the tenth formula (10): Among them, loc target represents position information, f2 represents nonlinear function, represents the classification parameter, Represents the sparse coefficient.

7. A metasurface positioning device based on nonlinear frame dictionary learning, wherein the metasurface positioning device is implemented based on a metasurface device, a plurality of signal receiving devices, and a plurality of signal transmitting devices, characterized in that: include: Positioning data acquisition module: used to set the signal receiving device, signal transmitting device and metasurface device around the target area, adjust the reflection angle of the metasurface device by turning on each signal transmitting device in turn, and obtain the reflection angle data of the metasurface device relative to each signal transmitting device; Obtain the positioning data of the training target in the target area with the label matrix; and obtaining signal data of the target to be located; Calculation module: used to construct a dictionary learning model and a target classification model by combining the label matrix method and the dictionary learning method, and add a nonlinear function based on the combination of the dictionary learning model and the target classification model to calculate the discriminant dictionary of the positioning data and the classification parameters in the target area; Sparse coefficient acquisition module: used to obtain the sparse coefficients of the target to be located by using a sparse coding method based on the discriminant dictionary; Position information acquisition module: used to obtain the position information of the target to be located based on the sparse coefficient and the classification parameter.

8. An electronic device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images