Wireless identity authentication method and system based on sparse attention mask

Abstract

The invention provides a wireless identity authentication method and system based on a sparse attention mask, which can be applied to the technical fields of signal processing, artificial intelligence and network security. The method comprises the following steps: representing an original radio frequency data sample as a 4D signal tensor sample; obtaining a target information dense region set by using the 4D signal tensor sample; obtaining a 4D signal tensor sample shielded by a mask by using the target information dense region; performing self-recovery iteration pre-training on a sparse encoder-decoder module by using the 4D signal tensor sample subjected to mask shielding; taking a pre-trained encoder as an encoder of the identity authentication model; performing iterative training on the identity authentication model by using the wireless training data set to obtain a trained identity authentication model; and processing the wireless radio frequency signal of the target area by using the trained identity authentication model to obtain an identity authentication result.

Description

Technical Field

[0001] The present invention relates to the technical fields of signal processing, artificial intelligence and network security, and in particular to a wireless identity authentication method and system based on a sparse attention mask, an electronic device and a storage medium. Background Art

[0002] At present, identity authentication has become an important research direction in the field of network security. Different from traditional authentication methods such as fingerprint or vision, wireless signals have the characteristics of non-contact operation, natural penetration and high privacy in cyberspace. Based on these advantages, wireless sensing technology has shown great application potential in the direction of secure identity authentication, and is used to explore the potential completion of secure identity authentication tasks in a more covert way in actual scenarios.

[0003] Wireless identity authentication can use signals to sense the unique behavioral information or physiological information of different people, and extract individual unique features from them, and finally perform secure identity authentication. However, there are many technical problems with existing technical solutions. For example, wireless identity authentication is difficult to physically model in actual environments, and it is usually necessary to combine deep learning technology to meet more complex perception requirements such as obtaining identity features. In addition, wireless signal data is different from visual or text, which is difficult to understand intuitively and difficult to label, resulting in learning-based methods being easily limited by the scale of labeled data. Therefore, under limited data samples, existing learning-based wireless identity authentication technical solutions lack generalization and robustness in practical applications. Summary of the invention

[0004] In view of the above technical problems, the present invention provides a wireless identity authentication method and system based on sparse attention mask, a method and system, an electronic device and a storage medium, which are used to solve at least one of the above technical problems.

[0005] According to a first aspect of the present invention, there is provided a wireless identity authentication method based on a sparse attention mask, comprising:

[0006] Using signal processing technology to characterize the original RF data samples into 4D signal tensor samples;

[0007] A preliminary set of dense regions is randomly generated on the 4D signal tensor samples, and the preliminary set of dense regions is sorted and screened to obtain a target information dense region set, and a sparse attention mask is generated for the 4D signal tensor samples using the target information dense region set to obtain the masked 4D signal tensor samples;

[0008] The sparse encoder-decoder module is self-recoveringly pre-trained using the masked 4D signal tensor samples and the preset reconstruction loss function to obtain a pre-trained encoder, and the pre-trained encoder is used as the encoder of the identity authentication model;

[0009] Iteratively train the identity authentication model using a wireless training data set with human contour labels and posture labels and an identity authentication loss function to obtain a trained identity authentication model;

[0010] The trained identity authentication model is used to process the wireless radio frequency signal in the target area to obtain the identity authentication result of the person in the target area.

[0011] According to an embodiment of the present invention, the above-mentioned characterization of the original RF data samples into 4D signal tensor samples by using signal processing technology includes:

[0012] Performing signal processing on the original radio frequency data samples to obtain the arrival angle-flight time and Doppler frequency shift of the original radio frequency data samples, wherein the original radio frequency data samples are the radio frequency data of the dynamic targets in the experimental area;

[0013] Using the amplitude image of the original RF data sample as the reconstruction target, perform vector concatenation of the arrival angle-flight time and Doppler frequency shift based on the time dimension to obtain a 4D signal tensor sample; or

[0014] The amplitude image of the original RF data sample is used as the reconstruction target, and the arrival angle-flight time is processed to obtain a 4D signal tensor sample.

[0015] According to an embodiment of the present invention, the above-mentioned sorting and screening of the preliminary set of dense areas to obtain the target information dense area set includes:

[0016] Based on the correspondence between the signal high energy area and the information density, the areas in the dense area preliminary selection set are sorted based on the sum of signal energy to obtain a sorted dense area preliminary selection set;

[0017] The sorted preliminary set of dense regions is screened based on a greedy strategy, and multiple non-overlapping regions with the highest ranking in the preliminary set of dense regions are selected to construct a target information dense region set.

[0018] According to an embodiment of the present invention, the generation of a sparse attention mask for a 4D signal tensor sample using a target information dense region set to obtain a masked 4D signal tensor sample includes:

[0019] A preset mask ratio is applied to each region in the target information dense region set, and regions in the 4D signal tensor samples that do not belong to the target information dense region set are masked to obtain masked 4D signal tensor samples.

[0020] According to an embodiment of the present invention, the sparse encoder-decoder module is self-recoveringly iteratively pre-trained using the masked 4D signal tensor samples and the preset reconstruction loss function to obtain the pre-trained encoder, including:

[0021] Construct a sparse encoder-decoder module for character feature processing, and improve the reconstruction loss function to obtain a preset reconstruction loss function;

[0022] Using the encoder of the sparse encoder-decoder module to extract semantic features from the masked 4D signal tensor samples, obtaining semantic feature samples;

[0023] The decoder of the sparse encoder-decoder module is used to map and reconstruct the semantic feature samples to obtain the 4D signal tensor samples after the mask is self-recovered;

[0024] The 4D signal tensor samples and the masked self-recovered 4D signal tensor samples are processed using a preset reconstruction loss function to obtain a reconstruction loss value;

[0025] The sparse encoder-decoder module is updated with the reconstruction loss value to obtain a sparse encoder-decoder module with updated parameters;

[0026] The semantic feature extraction operation, the mapping reconstruction operation, the reconstruction loss value calculation operation, and the model parameter update operation are iterated until the first preset training condition is met to obtain a pre-trained encoder.

[0027] According to an embodiment of the present invention, the preset reconstruction loss function is shown in the following formula:

[0028] ,

[0029] in, represents the target information dense area set, represents the decoder in the sparse encoder-decoder module, represents the encoder in the sparse encoder-decoder module, Represents the mask corresponding to the target information dense area set, represents the L2 norm.

[0030] According to an embodiment of the present invention, the above-mentioned wireless training data set with human body contour labels and posture labels and the identity authentication loss function are used to iteratively train the identity authentication model, and the trained identity authentication model is obtained, which includes:

[0031] Based on the preset weights, the identity classification loss function, the human body unique posture information estimation error loss function and the human body unique contour information generation loss function are weighted to obtain a preset identity authentication loss function;

[0032] The 4D signal tensor samples are processed using an identity authentication model based on a pre-trained encoder to obtain an identity authentication result of a dynamic experimental target, wherein the identity authentication result of the dynamic experimental target includes human body contour information and posture information;

[0033] The preset identity authentication loss function is used to process the identity authentication result of the dynamic experimental target and the human body contour label and posture label of the dynamic experimental target in the wireless data sample to obtain the identity authentication loss value;

[0034] The identity authentication model is updated with the identity authentication loss value to obtain an identity authentication model with updated parameters.

[0035] The model data processing operation, the loss value calculation operation and the model parameter update operation are iterated until the second preset training condition is met to obtain a trained identity authentication model.

[0036] A second aspect of the present invention provides a wireless identity authentication system based on a sparse attention mask, comprising:

[0037] A signal characterization module, used for characterizing the original RF data samples into 4D signal tensor samples using signal processing technology;

[0038] A sparse attention mask module is used to randomly generate a preliminary set of dense regions on 4D signal tensor samples, sort and filter the preliminary set of dense regions to obtain a target information dense region set, and use the target information dense region set to generate a sparse attention mask for the 4D signal tensor samples to obtain the masked 4D signal tensor samples;

[0039] A self-recovery pre-training module is used to perform self-recovery iterative pre-training on the sparse encoder-decoder module using the masked 4D signal tensor samples and a preset reconstruction loss function to obtain a pre-trained encoder, and use the pre-trained encoder as the encoder of the identity authentication model;

[0040] The identity authentication model training module is used to iteratively train the identity authentication model using a wireless training data set with human body contour labels and posture labels and an identity authentication loss function to obtain a trained identity authentication model;

[0041] The wireless identity authentication module is used to process the wireless radio frequency signal of the target area using the trained identity authentication model to obtain the identity authentication result of the person in the target area.

[0042] A third aspect of the present invention provides an electronic device, comprising: one or more processors; a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0043] The fourth aspect of the present invention further provides a computer-readable storage medium on which a computer program or instruction is stored, and the steps of the above method are implemented when the above computer program or instruction is executed by a processor.

[0044] The wireless identity authentication method based on sparse attention mask provided by the present invention is a new sparse perceptual masked autoencoder framework for radio frequency sensing tasks; a pre-training framework for obtaining general semantic representations from large-scale radio frequency signal data sets using self-supervised learning based on (Masked AutoEncoder) MAE is used, and the ability to obtain unique behavioral characteristics of the human body is particularly targeted. In addition, the present invention proposes a sparse perceptual masking strategy that can effectively bridge the information gap in radio frequency signals. By focusing on small areas with dense information and discarding areas with less information, not only the radio frequency sensing performance is improved, but also the computing and memory resource consumption are optimized. At the same time, the present invention redesigns the above-mentioned sparse encoder-decoder module to use a high-dimensional RF data cube, improves the processing performance of the identity authentication model for RF signals, saves storage space and speeds up the calculation speed; and because the inherent sparsity of RF signals is taken into account, a sparse masking scheme is provided, thereby significantly improving the ability of the model pre-training to extract individual characteristics of personnel, thereby showing the actual authentication advantages of higher accuracy and stronger generalization in the identity authentication model. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The above contents and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings, in which:

[0046] Figure 1 is an application scenario diagram of a wireless identity authentication method based on a sparse attention mask according to an embodiment of the present invention;

[0047] Figure 2 is a flow chart of a wireless identity authentication method based on sparse attention mask according to an embodiment of the present invention;

[0048] Figure 3 is a training framework diagram of a sparse encoder-decoder module and an identity authentication model according to an embodiment of the present invention;

[0049] Figure 4 is a schematic diagram of a process of generating a sparse attention mask according to an embodiment of the present invention;

[0050] Figure 5is a schematic diagram of a pre-training framework of a sparse encoder-decoder module according to an embodiment of the present invention;

[0051] Figure 6 It is a shared encoder block diagram of the pre-training framework and the identity authentication model according to an embodiment of the present invention;

[0052] Figure 7 is a comparative experimental effect diagram of a wireless identity authentication method according to an embodiment of the present invention;

[0053] Figure 8 is a structural block diagram of a wireless identity authentication system based on sparse attention mask according to an embodiment of the present invention;

[0054] Fig. 9 4 is a block diagram of an electronic device suitable for implementing a wireless identity authentication method based on a sparse attention mask according to an embodiment of the present invention. DETAILED DESCRIPTION

[0055] Below, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of concepts of the present invention.

[0056] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "comprise", "include", etc. used herein indicate the existence of the features, steps, operations and / or components, but do not exclude the existence or addition of one or more other features, steps, operations or components.

[0057] All terms (including technical and scientific terms) used herein have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0058] When using expressions such as "at least one of A, B, and C, etc.", they should generally be interpreted according to the meaning of the expression commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0059] In order to cope with the existing challenges of wireless identity authentication tasks, it is necessary to use self-supervised pre-training methods to enhance the model's ability to perceive individual identity features from large-scale unlabeled data, and ultimately achieve the purpose of greatly improving the accuracy, generalization and robustness of wireless identity authentication, better promote large-scale wireless perception applications represented by wireless identity authentication, and show its good development prospects in the fields of network security and wireless perception. Therefore, the present invention designs an efficient self-supervised pre-training framework based on masked autoencoders specific to wireless signals, and uses this framework to perform efficient self-supervised learning on large-scale unlabeled wireless signal data, thereby obtaining a codec that has been pre-trained to learn unique human features, and further applies it to the identity authentication model. The model directly generates a person's unique motion posture and body contour from the input radar signal, and performs identity authentication and recognition from the individual's subtle body differences and movement habit differences. With the help of pre-trained prior knowledge, the model can obtain a higher identity authentication accuracy rate and stronger robustness in different environments and different people by fine-tuning with only a small amount of labeled data. Obviously, this contactless method is more capable of completing identity authentication tasks, and the efficient self-supervised pre-training framework of the masked autoencoder proposed in the present invention can obtain deep unique human features at low cost, improve the accuracy and generalization of identity authentication, and show greater application value in actual scenarios.

[0060] Figure 1 4 is an application scenario diagram of a wireless identity authentication method based on a sparse attention mask according to an embodiment of the present invention.

[0061] like Figure 1 As shown, the application scenario 100 according to this embodiment may include signal processing, artificial intelligence, network security, etc. The network 104 is used to provide a medium for a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0062] The user can use the first terminal device 101, the second terminal device 102, and the third terminal device 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, the second terminal device 102, and the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only for example).

[0063] The first terminal device 101, the second terminal device 102, and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0064] The server 105 may be a server that provides various services, such as a background management server (only as an example) that provides support for websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process the received data such as user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal device.

[0065] It should be noted that the wireless identity authentication method based on the sparse attention mask provided in the embodiment of the present invention can generally be executed by the server 105. Accordingly, the wireless identity authentication system based on the sparse attention mask provided in the embodiment of the present invention can generally be set in the server 105. The wireless identity authentication method based on the sparse attention mask provided in the embodiment of the present invention can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the wireless identity authentication system based on the sparse attention mask provided in the embodiment of the present invention can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0066] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0067] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0068] At the same time, in the technical solution of the present invention, the user information involved (including but not limited to user personal information, user image information, user device information, such as location information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0069] In addition, in the scenario of using personal information for automated decision-making, the methods, devices, and systems provided by the embodiments of the present invention provide users with corresponding operation portals for users to choose to agree or reject the automated decision-making results; if the user chooses to reject, the expert decision-making process will be entered. The expression "automated decision-making" here refers to the activity of automatically analyzing and evaluating a person's behavioral habits, interests and hobbies, or economic, health, credit status, etc. through computer programs, and making decisions. The expression "expert decision-making" here refers to the activity of making decisions by people who specialize in a certain field, have specialized experience, knowledge and skills, and have reached a certain level of professionalism.

[0070] The following will be based on Figure 1 The scene described by Figure 2~Figure 7 The wireless identity authentication method based on sparse attention mask of the disclosed embodiment is described in detail.

[0071] Figure 2 4 is a flow chart of a wireless identity authentication method based on sparse attention mask according to an embodiment of the present invention.

[0072] like Figure 2 As shown, the wireless identity authentication method based on sparse attention mask of this embodiment includes operation S210 to operation S250.

[0073] In operation S210 , the original radio frequency data samples are characterized as 4D signal tensor samples using signal processing technology.

[0074] The above-mentioned original radio frequency data samples may be wireless WiFi signals or radar signals.

[0075] By performing signal processing on the above raw radio frequency data samples, characteristic information of the dynamic experimental target in a specific experimental area, such as the position of the dynamic experimental target, etc., can be obtained.

[0076] In operation S220, a preliminary set of dense regions is randomly generated on the 4D signal tensor samples, the preliminary set of dense regions is sorted and screened to obtain a target information dense region set, and a sparse attention mask is generated for the 4D signal tensor samples using the target information dense region set to obtain the masked 4D signal tensor samples.

[0077] The above operation S220 performs masking through a sparse mask, so that relevant technical personnel can focus more on the information-intensive area of ​​the dynamic experimental target. This method can greatly reduce the consumption of computing and memory resources.

[0078] In operation S230, the sparse encoder-decoder module is self-recoveringly iteratively pre-trained using the masked 4D signal tensor samples and a preset reconstruction loss function to obtain a pre-trained encoder, and the pre-trained encoder is used as an encoder of the identity authentication model.

[0079] The sparse encoder-decoder module can self-recover the mask in the masked 4D signal tensor samples to obtain the masked self-recovered 4D signal tensor samples.

[0080] The above-mentioned sparse encoder-decoder is redesigned according to the characteristics of RF data to reduce the consumption of computing and memory resources and improve the generalization of the pre-trained encoder.

[0081] In operation S240, the identity authentication model is iteratively trained using a wireless training data set having a human body contour label and a posture label and an identity authentication loss function to obtain a trained identity authentication model.

[0082] The encoder of the identity authentication model uses the encoder trained in operation S230.

[0083] In operation S250, the trained identity authentication model is used to process the radio frequency signal of the target area to obtain an identity authentication result of the person in the target area.

[0084] In an embodiment of the present invention, before obtaining the user's information, the user's consent or authorization may be obtained. For example, before operation S250, a request to obtain the user's information may be issued to the user. When the user agrees or authorizes that the user's information may be obtained, the operation S250 is performed.

[0085] The wireless identity authentication method based on sparse attention mask provided by the present invention is a new sparse perceptual masked autoencoder framework for radio frequency sensing tasks; a pre-training framework for obtaining general semantic representations from large-scale radio frequency signal data sets using self-supervised learning based on (Masked AutoEncoder) MAE is used, and the ability to obtain unique behavioral characteristics of the human body is particularly targeted. In addition, the present invention proposes a sparse perceptual masking strategy that can effectively bridge the information gap in radio frequency signals. By focusing on small areas with dense information and discarding areas with less information, not only the radio frequency sensing performance is improved, but also the computing and memory resource consumption are optimized. At the same time, the present invention redesigns the above-mentioned sparse encoder-decoder module to use a high-dimensional RF data cube, improves the processing performance of the identity authentication model for RF signals, saves storage space and speeds up the calculation speed; and because the inherent sparsity of RF signals is taken into account, a sparse masking scheme is provided, thereby significantly improving the ability of the model pre-training to extract individual characteristics of personnel, thereby showing the actual authentication advantages of higher accuracy and stronger generalization in the identity authentication model.

[0086] The following is a specific implementation method and combined with the attached Figure 3 The architecture of the wireless identity authentication method based on sparse attention mask provided by the present invention is further described in detail.

[0087] Figure 3 It is a training framework diagram of a sparse encoder-decoder module and an identity authentication model according to an embodiment of the present invention.

[0088] like Figure 3 As shown in the figure, the present invention first designs a universal masked autoencoder pre-training framework specific to wireless signals. Given an RF (Radio Frequency) signal frame, a signal processing algorithm is used to create a signal representation consistent with the requirements of the downstream RF sensing task. The signal representation is processed by sparse attention masking to obtain its masked version. The masked signal representation is input into the encoder backbone to obtain a high-level semantic representation. At the same time, the decoder attempts to predict the masked part. Therefore, the goal of reconstruction is to capture the high-quality unique individual information representation contained in the RF signal by minimizing the loss. After an efficient pre-training process, the encoder that retains the deep individual unique behavioral features is directly applied to the identity authentication model training to greatly improve the accuracy of identity authentication. Specifically, the present invention is mainly implemented in the following five technical modules, namely: (1) signal representation; (2) sparse attention masking; (3) encoder-decoder construction; (4) reconstruction loss; (5) identity authentication model training module.

[0089] According to an embodiment of the present invention, the above-mentioned use of signal processing technology to characterize the original RF data samples as 4D signal tensor samples includes: performing signal processing on the original RF data samples to obtain the arrival angle-flight time and Doppler frequency shift of the original RF data samples, wherein the original RF data samples are wireless RF data of dynamic targets in the experimental area; using the amplitude image of the original RF data samples as the reconstruction target, performing vector splicing based on the time dimension on the arrival angle-flight time and Doppler frequency shift to obtain 4D signal tensor samples; or using the amplitude image of the original RF data samples as the reconstruction target, processing the arrival angle-flight time, and obtaining 4D signal tensor samples.

[0090] In the process of acquiring 4D signal tensor samples, the arrival angle-flight time and Doppler frequency shift can be spliced ​​and acquired; or only the arrival angle-flight time can be used for acquisition.

[0091] The process of obtaining the 4D signal tensor samples is further described in detail below through a specific implementation method.

[0092] Given raw RF data, such as radar signals, the present invention can use various mature signal processing techniques to obtain their corresponding signal representations, such as Angle of Arrival (AoA)-Time of Flight (ToF) and Doppler Frequency Shift (DFS)).

[0093] Among them, the arrival angle-flight time determines the angle of signal arrival through a multi-antenna array to estimate the direction, and the distance is estimated by the time difference of the signal from the base station to the mobile object and then back. It can be expressed by formula (1):

[0094] (1),

[0095] in It's AoA. It’s ToF. is the spatial separation between two adjacent antennas, is the signal frequency, is the speed of light, is the difference between adjacent frequencies, is the receiving signal, and is the index of the receiver antenna and the signal frequency. The output of AoA-ToF is a 2D matrix, please note that the matrices obtained from radar and WiFi are usually significantly different due to the difference in resolution.

[0096] Among them, Doppler frequency shift (DFS) indicates that the movement of the target causes the change in the length of the signal path reflected by the target, thereby generating a certain frequency offset in the observed signal. By performing time-frequency analysis on the CSI power, the present invention can extract the DFS caused by the target movement from the dynamic change of the power. The Doppler information is extracted by applying conjugate division, principal component analysis (PCA) algorithm and short-time Fourier transform (STFT) to the CSI data. The output of DFS is a dimension of A frequency-time matrix, where represents the frequency domain, Represents the time domain.

[0097] However, compared with visual images and language, the signal representation of most RF signals is globally sparse and locally dense, that is, the information carried by the RF signal is concentrated in a few areas, which makes MAE-based RF pre-training difficult. In addition, different signal representations have different characteristics (such as dimensions, features, etc.), for example, AoA-ToF is a 3D matrix (Time-AoA-ToF), while DFS is a 2D matrix (frequency-time). For simplicity, the signal representation is abstracted into a general 4D tensor form. In particular, the present invention does not process the spectral image as a conventional real-valued RGB image, because this will cause the loss of finer phase information in the signal. For example, the channel of AOA-TOF is 2, including real and imaginary parts, instead of taking the modulus and using the amplitude image as input. However, in the generation process, the amplitude image is used as the reconstruction target because it is found that this trick accelerates the convergence process. The input tensor is divided into a set of small blocks, each of which contains a representation tensor. The present invention aims to use small blocks with dense information in part to predict other small blocks that are masked, so as to obtain a high-level understanding of the RF signal.

[0098] According to an embodiment of the present invention, the above-mentioned sorting and screening of the preliminary set of dense areas to obtain the target information dense area set includes: based on the correspondence between the signal high-energy area and the information density, the areas in the preliminary set of dense areas are sorted based on the sum of signal energies to obtain the sorted preliminary set of dense areas; based on the greedy strategy, the sorted preliminary set of dense areas is screened, and multiple non-overlapping areas with top rankings in the sorted preliminary set of dense areas are selected to construct the target information dense area set.

[0099] According to an embodiment of the present invention, the above-mentioned generation of a sparse attention mask for 4D signal tensor samples using a target information dense area set to obtain a masked 4D signal tensor sample includes: applying a preset mask ratio to each area in the target information dense area set, and masking the areas in the 4D signal tensor sample that do not belong to the target information dense area set to obtain the masked 4D signal tensor sample.

[0100] The following is a specific implementation method and combined with the attached Figure 4 The process of obtaining the 4D signal tensor samples after masking provided by the present invention is further described in detail.

[0101] Figure 4 is a schematic diagram of a process of generating a sparse attention mask according to an embodiment of the present invention.

[0102] In order to solve the problem of RF signals with sparse global information and dense local information, the present invention proposes a customized masking strategy FOCUS. Specifically, the present invention dynamically tracks information-dense areas and only retains some patches in these information-dense areas, while discarding all other patches. In this way, the pre-training process can observe some information-dense patches and try to predict other patches to gain a deep understanding of the RF signal representation. In addition, many information-sparse patches do not participate in model training, which greatly reduces the consumption of computing and memory resources.

[0103] The FOCUS masking process mainly includes the following two steps: Figure 4 As shown:

[0104] Information dense area selection: The present invention first randomly generates a preliminary set of dense areas, expressed as , where the primaries can be generated across the entire region and contain patches, including and Then, since high-energy regions are usually information-dense, the region proposal set is selected based on the sum of signal energies. Follow the greedy strategy and select the first non-overlapping regions as the target information dense region set ;

[0105] Mask generation: The present invention applies mask ratio only to each information-intensive area , and mask other blocks, including patches in information-sparse regions. Overall, this is equivalent to applying a larger masking ratio Applied to the entire region to obtain a masked version of the signal representation.

[0106] According to an embodiment of the present invention, the above-mentioned self-recovery iterative pre-training of the sparse encoder-decoder module using the masked 4D signal tensor samples and the preset reconstruction loss function to obtain the pre-trained encoder includes: constructing a sparse encoder-decoder module for character feature processing, and improving the reconstruction loss function to obtain a preset reconstruction loss function; using the encoder of the sparse encoder-decoder module to extract semantic features from the masked 4D signal tensor samples to obtain semantic feature samples; using the decoder of the sparse encoder-decoder module to map and reconstruct the semantic feature samples to obtain the masked self-recovered 4D signal tensor samples; using the preset reconstruction loss function to process the 4D signal tensor samples and the masked self-recovered 4D signal tensor samples to obtain a reconstruction loss value; using the reconstruction loss value to update the parameters of the sparse encoder-decoder module to obtain the sparse encoder-decoder module after the parameter update; iteratively performing semantic feature extraction operations, mapping and reconstruction operations, reconstruction loss value calculation operations and model parameter update operations until the first preset training condition is met to obtain the pre-trained encoder.

[0107] The following is a specific implementation method and combined with the attached Figure 5 and 6 The training process of the sparse encoder-decoder module provided by the present invention is further described in detail.

[0108] Figure 5 2 is a schematic diagram of a pre-training framework of a sparse encoder-decoder module according to an embodiment of the present invention.

[0109] Figure 6 It is a block diagram of an encoder shared by a pre-training framework and an identity authentication model according to an embodiment of the present invention.

[0110] like Figure 5 As shown, the goal of the encoder is to convert the mask signal representation into a high semantic representation, which can be any classic convolutional network such as ResNet or ConvNeXt. However, compared with ordinary convolutional networks, the present invention uses a sparse version of the convolutional layer. Sparse convolution not only reduces the computational resource consumption in the mask patch, but more importantly, avoids information leakage from the observed patch to its neighboring mask patches and representation collapse caused by ordinary convolution calculations in many zero-valued mask regions.

[0111] The role of the decoder is to map the high semantic representation to the reconstructed masked patch. Therefore, the present invention adopts a lightweight design of a common convolutional decoder, allowing communication between information-intensive patches and masked patches, while forcing the encoder to have a high-level understanding, rather than simply transmitting the input to the decoder and causing the representation to collapse. The lightweight decoder also improves the efficiency of pre-training. In addition, it should be noted that the decoder is only used in the pre-training stage. Once the pre-training is completed, the decoder will be discarded and replaced with a different sub-task network adapted to the purpose of the downstream task.

[0112] In terms of reconstruction loss function: the framework based on classic mask pre-training focuses on restoring all mask regions. Although the framework of the present invention also restores all mask patches, the difference is that PRISM mainly focuses on restoring mask patches in information-dense areas.

[0113] According to an embodiment of the present invention, the preset reconstruction loss function is as shown in formula (2):

[0114] (2),

[0115] in, represents the target information dense area set, represents the decoder in the sparse encoder-decoder module, represents the encoder in the sparse encoder-decoder module, Represents the mask corresponding to the target information dense area set, represents the L2 norm.

[0116] At the same time, the present invention redesigns the original encoder architecture to adapt to the high-dimensional RF data cube, such as Figure 6 As shown, it involves optimizing the sampling rate, adjusting the convolution kernel size, and transforming the spatiotemporal separation convolution, which are all expected to synergistically improve the model's processing performance for RF signals, save storage space, and speed up calculations. In particular, it is necessary to design a new masking strategy taking into account the inherent sparsity of RF signals, paying special attention to meaningful sparse areas, thereby significantly improving the model's ability to pre-train and extract individual features of people, thereby showing actual authentication advantages of higher accuracy and stronger generalization in the identity authentication model.

[0117] According to an embodiment of the present invention, the above-mentioned wireless training data set with human body contour labels and posture labels and the identity authentication loss function are used to iteratively train the identity authentication model to obtain the trained identity authentication model, including: based on preset weights, the identity classification loss function, the human body unique posture information estimation error loss function and the human body unique contour information generation loss function are weighted to obtain a preset identity authentication loss function; the 4D signal tensor sample is processed using the identity authentication model based on the pre-trained encoder to obtain the identity authentication result of the dynamic experimental target, wherein the identity authentication result of the dynamic experimental target includes human body contour information and posture information; the identity authentication result of the dynamic experimental target and the human body contour label and posture label of the dynamic experimental target in the wireless data sample are processed using the preset identity authentication loss function to obtain the identity authentication loss value; the identity authentication loss value is used to update the parameters of the identity authentication model to obtain the identity authentication model after the parameter update; the model data processing operation, the loss value calculation operation and the model parameter update operation are iteratively performed until the second preset training condition is met to obtain the trained identity authentication model.

[0118] The training process of the above-mentioned identity authentication model is further explained in detail below through a specific implementation method.

[0119] The encoder of the identity authentication model is shared with the pre-trained model to directly obtain the prior of the unique information of the human body. The present invention mainly performs identity authentication and recognition based on the contour and posture information of the human body, so the identity authentication network requires the human contour and posture features generated by the model to assist in identity recognition. Therefore, the present invention simultaneously retains the above-mentioned human contour segmentation and 3D posture estimation intermediate task network and the corresponding output results, and combines the identity classification network constructed by the linear layer. The loss function of the former two is added to the identity classification loss as an auxiliary loss to ensure effective feature extraction in the model learning process and improve the classification loss accuracy. The loss function is shown in formula (3):

[0120] (3),

[0121] in, is the identity classification loss, is the error loss of estimating the unique posture information of the human body, Generate loss for the unique contour information of the human body, are the preset weights of the three losses, Total training loss for the final authentication model.

[0122] In order to better illustrate the advantages and effectiveness of the above method provided by the present invention, the following specific experiments are combined with the attached Figure 7 The above method provided by the present invention is further verified.

[0123] Figure 7 It is a comparative experimental effect diagram of the wireless identity authentication method according to an embodiment of the present invention.

[0124] In the specific experimental setting, the present invention uses the radar-based open source dataset HIBER, which includes various environments, users, occlusions and actions. It is collected by capturing RF signals from horizontal and vertical planes using two vertical FMCW radars. The dataset provides RF heatmaps, RGB images, 2D / 3D skeletons, bounding boxes and contour ground truth.

[0125] In addition, the present invention uses the masked autoencoder pre-training framework proposed in the present invention to enhance the identity authentication model to better extract relevant unique identity information such as human body contour or posture. Furthermore, the present invention needs to verify the feasibility of identity authentication based on Radar-perceived information such as human body contour and posture, and that the PRISM framework can enhance the accuracy of identity authentication. For this purpose, the present invention also selects view2, 3, 4, 5, 7, and 9 in the data set as pre-training samples, totaling 218,650 samples.

[0126] like Figure 7 As shown in Figure 1, the present invention first tests the information extraction effect of the pre-training framework on the model to extract the unique behavioral characteristics of the human body. For contour segmentation, the present invention uses view6 for training (17,520 samples) and testing (4664 samples). For 3D pose estimation, the present invention uses view Figure 1 , 6 , 8, and 10 for training (32584 samples) and testing (8584 samples). The experiments show that the IoU (Intersection of Union) of the model trained from scratch for extracting unique contour features is 0.7444, 0.7485, and 0.7473, respectively, which is 0.0425, 0.0384, and 0.0413 higher than the classic supervised baseline. After the pre-training framework proposed in the present invention, the IoU of each backbone is 0.7642, 0.7635, and 0.7630, respectively, which is significantly improved by 0.0198, 0.015, and 0.0157 compared with their respective supervised training. At the same time, after pre-training, the effect of extracting unique posture features is compared, and the Euclidean distance error of the joint point estimation of each backbone network in 3D posture estimation is improved by 23.79mm, 25.44mm, and 19.32mm, respectively. In addition, the overall best result is 72.67mm.

[0127] According to the above, after the model is pre-trained with the masked autoencoder proposed by the present invention, the ability to extract unique human behavior features (contour or posture) is significantly improved, and the recognition accuracy of identity authentication can be further improved. For the training set and test set of the downstream task, in order to ensure fairness, the present invention first maintains the same division method as the posture estimation task, and at the same time uses view 1, 6, 8, and 10 as test environments for k-fold cross validation. After model training and testing, the final identity recognition accuracy results show that when the same backbone network is used, the network trained from scratch can only achieve an identity authentication accuracy of about 60% in 4 environments, and the overall average accuracy is 61.2%. However, after extracting more accurate human contour and posture information through the pre-training framework, the identity authentication accuracy can reach 75% in 4 environments, and the overall average accuracy is 75.7%, which significantly improves the accuracy by 14.5%. Therefore, it can be shown through experiments that the pre-trained enhanced model can extract more accurate individual identity feature information such as human contour and posture to obtain a higher identity authentication accuracy.

[0128] In addition, the present invention further tests the identity verification accuracy results under three different external conditions (normal lighting unobstructed environment, obstructed environment, and dark environment), such as Figure 7 As shown in the figure, the network trained from scratch can only achieve 60.4%, 51.2%, and 57.6% authentication accuracy under three conditions. However, after the pre-trained network extracts more accurate human body contour and posture information, it can achieve 77.6%, 66.1%, and 75.3% authentication accuracy under three conditions. In the end, the accuracy was significantly improved by 17.2%, 14.9%, and 17.7%, respectively, which fully highlights the significant improvement of the model after pre-training with the PRISM framework, and also demonstrates the potential advantages and development prospects of wireless signals for authentication in special scenarios.

[0129] Based on the above-mentioned wireless identity authentication method based on sparse attention mask, the present invention also provides a wireless identity authentication system based on sparse attention mask. Figure 8 The system is described in detail.

[0130] Figure 8 4 is a structural block diagram of a wireless identity authentication system based on sparse attention mask according to an embodiment of the present invention.

[0131] like Figure 8 As shown, the wireless identity authentication system 800 based on sparse attention mask includes a signal characterization module 810 , a sparse attention mask module 820 , a self-recovery pre-training module 830 , an identity authentication model training module 840 and a wireless identity authentication module 850 .

[0132] The signal characterization module 810 is used to characterize the original RF data samples into 4D signal tensor samples by using signal processing technology. In one embodiment, the signal characterization module 810 can be used to perform the operation S210 described above, which will not be repeated here.

[0133] The sparse attention mask module 820 is used to randomly generate a preliminary set of dense areas on the 4D signal tensor samples, sort and filter the preliminary set of dense areas to obtain a target information dense area set, and use the target information dense area set to generate a sparse attention mask for the 4D signal tensor samples to obtain the masked 4D signal tensor samples; in one embodiment, the sparse attention mask module 820 can be used to perform the operation S220 described above, which will not be repeated here.

[0134] The self-recovery pre-training module 830 is used to self-recoverably iteratively pre-train the sparse encoder-decoder module using the masked 4D signal tensor samples and the preset reconstruction loss function to obtain a pre-trained encoder, and use the pre-trained encoder as the encoder of the identity authentication model; in one embodiment, the self-recovery pre-training module 830 can be used to perform the operation S230 described above, which will not be repeated here.

[0135] The identity authentication model training module 840 is used to iteratively train the identity authentication model using a wireless training data set with human body contour labels and posture labels and an identity authentication loss function to obtain a trained identity authentication model; in one embodiment, the identity authentication model training module 840 can be used to perform the operation S240 described above, which will not be repeated here.

[0136] The wireless identity authentication module 850 is used to process the wireless radio frequency signal of the target area using the trained identity authentication model to obtain the identity authentication result of the person in the target area; in one embodiment, the wireless identity authentication module 850 can be used to perform the operation S250 described above, which will not be repeated here.

[0137] According to an embodiment of the present invention, any multiple modules of the signal characterization module 810, the sparse attention mask module 820, the self-recovery pre-training module 830, the authentication model training module 840, and the wireless authentication module 850 can be combined in one module, or any one of the modules can be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present invention, at least one of the signal characterization module 810, the sparse attention mask module 820, the self-recovery pre-training module 830, the authentication model training module 840, and the wireless authentication module 850 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or can be implemented by hardware or firmware such as any other reasonable way of integrating or packaging the circuit, or implemented in any one of the three implementation methods of software, hardware, and firmware, or in any appropriate combination of any of them. Alternatively, at least one of the signal characterization module 810, the sparse attention mask module 820, the self-recovery pre-training module 830, the authentication model training module 840, and the wireless authentication module 850 may be at least partially implemented as a computer program module, which may perform corresponding functions when the computer program module is executed.

[0138] Fig. 9 4 is a block diagram of an electronic device suitable for implementing a wireless identity authentication method based on a sparse attention mask according to an embodiment of the present invention.

[0139] like Fig. 9 As shown, the electronic device 900 according to an embodiment of the present invention includes a processor 901, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 902 or a program loaded from a storage part 908 to a random access memory (RAM) 903. The processor 901 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 901 may also include an onboard memory for caching purposes. The processor 901 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0140] In RAM 903, various programs and data required for the operation of electronic device 900 are stored. Processor 901, ROM 902 and RAM 903 are connected to each other via bus 904. Processor 901 performs various operations of the method flow according to the embodiment of the present invention by executing the program in ROM 902 and / or RAM 903. It should be noted that the program can also be stored in one or more memories other than ROM 902 and RAM 903. Processor 901 can also perform various operations of the method flow according to the embodiment of the present invention by executing the program stored in the one or more memories.

[0141] According to an embodiment of the present invention, the electronic device 900 may further include an input / output (I / O) interface 905, which is also connected to the bus 904. The electronic device 900 may further include one or more of the following components connected to the input / output (I / O) interface 905: an input portion 906 including a keyboard, a mouse, etc.; an output portion 907 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 908 including a hard disk, etc.; and a communication portion 909 including a network interface card such as a LAN card, a modem, etc. The communication portion 909 performs communication processing via a network such as the Internet. A drive 910 is also connected to the input / output (I / O) interface 905 as needed. A removable medium 911, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 910 as needed, so that a computer program read therefrom is installed into the storage portion 908 as needed.

[0142] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiment; or may exist independently without being assembled into the device / apparatus / system. The above computer-readable storage medium carries one or more programs, and when the above one or more programs are executed, the method according to the embodiment of the present invention is implemented.

[0143] According to an embodiment of the present invention, the computer-readable storage medium may be a non-volatile computer-readable storage medium, for example, it may include but is not limited to: 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), 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, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, an apparatus or a device. For example, according to an embodiment of the present invention, the computer-readable storage medium may include the ROM 902 and / or RAM 903 described above and / or one or more memories other than ROM 902 and RAM 903.

[0144] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, 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 above-mentioned module, program segment, or a part of a 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 different order from the order 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 or flow chart, and the combination of the boxes in the block diagram 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.

[0145] It will be appreciated by those skilled in the art that the features described in the various embodiments of the present invention may be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in the present invention. In particular, without departing from the spirit and teachings of the present invention, the features described in the various embodiments of the present invention may be combined and / or combined in various ways. All of these combinations and / or combinations fall within the scope of the present invention.

[0146] The embodiments of the present invention are described above. However, these embodiments are only for the purpose of illustration, and are not intended to limit the scope of the present invention. Although each embodiment is described above, it does not mean that the measures in each embodiment cannot be used in combination advantageously. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. A wireless identity authentication method based on sparse attention mask, characterized in that: The method comprises: Using signal processing technology to characterize the original RF data samples into 4D signal tensor samples; Randomly generating a preliminary set of dense regions on the 4D signal tensor samples, sorting and screening the preliminary set of dense regions to obtain a target information dense region set, and using the target information dense region set to generate a sparse attention mask for the 4D signal tensor samples to obtain a masked 4D signal tensor sample; Using the masked 4D signal tensor samples and a preset reconstruction loss function to self-recoveringly iterate pre-train a sparse encoder-decoder module to obtain a pre-trained encoder, and using the pre-trained encoder as an encoder of the identity authentication model; Iteratively training the identity authentication model using a wireless training data set with human body contour labels and posture labels and an identity authentication loss function to obtain a trained identity authentication model; The trained identity authentication model is used to process the wireless radio frequency signal of the target area to obtain the identity authentication result of the person in the target area.

2. The method according to claim 1, characterized in that The signal processing techniques used to characterize the raw RF data samples into 4D signal tensor samples include: Performing signal processing on the original radio frequency data samples to obtain the arrival angle-flight time and Doppler frequency shift of the original radio frequency data samples, wherein the original radio frequency data samples are the radio frequency data of the dynamic targets in the experimental area; Taking the amplitude image of the original RF data sample as a reconstruction target, performing vector concatenation based on the time dimension on the arrival angle-flight time and the Doppler frequency shift to obtain the 4D signal tensor sample; or The amplitude image of the original radio frequency data sample is used as a reconstruction target, and the arrival angle-flight time is processed to obtain the 4D signal tensor sample.

3. The method according to claim 1, characterized in that The preliminary set of dense areas is sorted and screened to obtain a target information dense area set including: Based on the correspondence between the signal high energy area and the information density, the areas in the dense area preliminary selection set are sorted based on the sum of signal energies to obtain a sorted dense area preliminary selection set; The sorted preliminary set of dense regions is screened based on a greedy strategy, and a plurality of non-overlapping regions with top rankings in the preliminary set of dense regions are selected to construct the target information dense region set.

4. The method according to claim 3, characterized in that Generating a sparse attention mask for the 4D signal tensor sample using the target information dense region set to obtain the masked 4D signal tensor sample includes: A preset mask ratio is applied to each region in the target information dense region set, and regions in the 4D signal tensor samples that do not belong to the target information dense region set are masked to obtain the masked 4D signal tensor samples.

5. The method according to claim 1, characterized in that: The sparse encoder-decoder module is self-recoveringly pre-trained iteratively using the masked 4D signal tensor samples and a preset reconstruction loss function, and the pre-trained encoder includes: Constructing a sparse encoder-decoder module for character feature processing, and improving the reconstruction loss function to obtain the preset reconstruction loss function; Using the encoder of the sparse encoder-decoder module to extract semantic features from the masked 4D signal tensor samples to obtain semantic feature samples; Using the decoder of the sparse encoder-decoder module to map and reconstruct the semantic feature samples to obtain 4D signal tensor samples after mask self-recovery; Processing the 4D signal tensor samples and the masked self-recovered 4D signal tensor samples using the preset reconstruction loss function to obtain a reconstruction loss value; Using the reconstruction loss value, the parameters of the sparse encoder-decoder module are updated to obtain a sparse encoder-decoder module after parameter update; The semantic feature extraction operation, the mapping reconstruction operation, the reconstruction loss value calculation operation, and the model parameter update operation are iterated until the first preset training condition is met to obtain the pre-trained encoder.

6. The method according to claim 5, characterized in that The preset reconstruction loss function is shown in the following formula: ， in, represents the target information dense area set, represents a decoder in the sparse encoder-decoder module, represents the encoder in the sparse encoder-decoder module, represents the mask corresponding to the target information dense area set, represents the L2 norm.

7. The method according to claim 1, characterized in that The identity authentication model is iteratively trained using a wireless training data set with human body contour labels and posture labels and an identity authentication loss function, and the trained identity authentication model includes: Based on the preset weights, the identity classification loss function, the human body unique posture information estimation error loss function and the human body unique contour information generation loss function are weighted to obtain the preset identity authentication loss function; Processing the 4D signal tensor samples using an identity authentication model based on the pre-trained encoder to obtain an identity authentication result of a dynamic experimental target, wherein the identity authentication result of the dynamic experimental target includes human body contour information and posture information; Using the preset identity authentication loss function to process the identity authentication result of the dynamic experiment target and the human body contour label and posture label of the dynamic experiment target in the wireless data sample to obtain an identity authentication loss value; Using the identity authentication loss value to update the parameters of the identity authentication model to obtain an identity authentication model with updated parameters; The model data processing operation, the loss value calculation operation and the model parameter update operation are iterated until the second preset training condition is met to obtain a trained identity authentication model.

8. A wireless identity authentication system based on sparse attention mask, characterized in that: The system comprises: A signal characterization module, used for characterizing the original RF data samples into 4D signal tensor samples using signal processing technology; A sparse attention mask module is used to randomly generate a preliminary set of dense regions on the 4D signal tensor samples, sort and screen the preliminary set of dense regions to obtain a target information dense region set, and use the target information dense region set to generate a sparse attention mask for the 4D signal tensor samples to obtain the masked 4D signal tensor samples; A self-recovery pre-training module, used to perform self-recovery iterative pre-training on a sparse encoder-decoder module using the masked 4D signal tensor samples and a preset reconstruction loss function to obtain a pre-trained encoder, and use the pre-trained encoder as an encoder of an identity authentication model; An identity authentication model training module, used to iteratively train the identity authentication model using a wireless training data set with human body contour labels and posture labels and an identity authentication loss function to obtain a trained identity authentication model; The wireless identity authentication module is used to process the wireless radio frequency signal of the target area by using the trained identity authentication model to obtain the identity authentication result of the person in the target area.

9. An electronic device, comprising: one or more processors; a memory for storing one or more instructions, Wherein, when the one or more instructions are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, enables the processor to implement the method according to any one of claims 1 to 7.

Images