Object authentication method, watermark embedding method and device of wireless sensing model

Abstract

The invention provides an object authentication method, a watermark embedding method and a watermark embedding device for a wireless sensing model, which can be applied to the technical field of signal processing. The method comprises the steps that an intermediate frequency receiving signal is obtained, and the intermediate frequency receiving signal is obtained by mixing a transmitting signal and an echo signal; performing phase offset compensation based on a fixed weight on the intermediate frequency receiving signal to generate a wireless signal spectrogram; the wireless signal spectrogram is input into a wireless sensing model, a target wireless signal spectrogram is output, the wireless sensing model comprises a plurality of feature extraction layers and is used for performing feature extraction on the wireless signal spectrogram, and at least one feature extraction layer is embedded with a pass authentication module and is used for performing pass authentication on the target wireless signal spectrogram; the pass authentication module is used for performing phase offset compensation based on dynamic weight on the current input so as to authenticate the pass of the target object; and determining an authentication result of the target object according to the target wireless signal spectrogram and a preset spectrogram.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and in particular to an object authentication method, a watermark embedding method and a device of a wireless perception model. Background Art

[0002] Wireless sensing refers to the process of sensing without interfering with the normal activities of the target object. The wireless sensing model is a model that uses the propagation characteristics of wireless signals, such as channel state information, to achieve sensing of the target object.

[0003] In the process of implementing the present invention, it is found that the related art mainly protects the wireless perception model by embedding watermarks, but the embedded watermarks are difficult to deal with the attacks of forged watermarks, resulting in low security of the wireless perception model. In addition, the wireless perception model can also be protected by judicial means after the fact, but the efficiency of post-processing is low, and it is difficult to achieve active protection of the wireless perception model. Summary of the invention

[0004] In view of the above problems, the present invention provides an object authentication method, a watermark embedding method and a device for a wireless perception model.

[0005] According to a first aspect of the present invention, there is provided an object authentication method of a wireless perception model, comprising: acquiring an intermediate frequency receiving signal, wherein the intermediate frequency receiving signal is obtained by mixing a transmitting signal and an echo signal; performing phase offset compensation based on a fixed weight on the intermediate frequency receiving signal to generate a wireless signal spectrum; inputting the wireless signal spectrum into a wireless perception model, and outputting a target wireless signal spectrum, wherein the wireless perception model comprises a plurality of feature extraction layers for performing feature extraction on the wireless signal spectrum, at least one feature extraction layer being embedded with a pass authentication module, the pass authentication module being used to perform phase offset compensation based on a dynamic weight on a current input to authenticate a pass of the target object; determining an authentication result of the target object according to the target wireless signal spectrum and a preset spectrum.

[0006] According to the object authentication method of the wireless perception model provided by the present invention, a wireless signal spectrum can be generated by performing fixed-weight phase offset compensation on the acquired intermediate frequency receiving signal, and then the wireless signal spectrum is processed by the wireless perception model to obtain a target wireless signal spectrum, and the authentication result of the target object is obtained by comparing the target wireless signal spectrum with a preset spectrum. Since at least one of the multiple feature extraction layers in the wireless perception model is embedded with a pass authentication module, the pass authentication module performs dynamic weight-based phase offset compensation on the current input, and the pass of the target object can be authenticated. When the quality of the target wireless signal spectrum is higher than the preset spectrum, the target object is an authorized object. When the quality of the target wireless signal spectrum is lower than the preset spectrum, the target object is an unauthorized object. Therefore, when the target object is an authorized object, the quality of the target wireless signal spectrum is improved, thereby improving the perception accuracy of the wireless perception model. Since the target object can be authenticated by the quality of the output target wireless signal spectrum, active protection of the wireless perception model is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] 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.

[0008] Figure 1 An application scenario diagram of an object authentication method based on a wireless perception model according to an embodiment of the present invention is shown.

[0009] Figure 2 A flow chart of an object authentication method of a wireless awareness model according to an embodiment of the present invention is shown.

[0010] Figure 3 The figure shows an architecture diagram of a pass authentication module according to an embodiment of the present invention.

[0011] Figure 4A A wireless signal spectrum diagram according to an embodiment of the present invention is shown.

[0012] Figure 4B A first target wireless signal spectrum diagram according to an embodiment of the present invention is shown.

[0013] Figure 4C A second target wireless signal spectrum diagram according to an embodiment of the present invention is shown.

[0014] Figure 5 A flow chart of a watermark embedding method of a wireless perception model according to an embodiment of the present invention is shown.

[0015] Figure 6 A diagram showing an example of generating a target character string watermark according to an embodiment of the present invention is shown.

[0016] Figure 7 An example diagram of generating a target pass according to an embodiment of the present invention is shown.

[0017] Figure 8 An example diagram of the perception process of the wireless perception model according to an embodiment of the present invention is shown.

[0018] Fig. 9A The figure shows an architecture diagram of a wireless awareness model according to an embodiment of the present invention.

[0019] Fig. 9B The standard modules of the wireless sensing model according to the embodiment of the present invention are shown.

[0020] Fig. 9C The protection module of the wireless awareness model according to the embodiment of the present invention is shown.

[0021] Fig.10 The sensing process of the wireless sensing model according to another embodiment of the present invention is shown.

[0022] Fig.11 A structural block diagram of an object authentication device based on a wireless perception model according to an embodiment of the present invention is shown.

[0023] Fig.12 A block diagram of an electronic device suitable for implementing an object authentication method of a wireless perception model according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0024] 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.

[0025] 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.

[0026] 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.

[0027] 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.).

[0028] In the technical solution of the present invention, the user information (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.) involved 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 the 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.

[0029] In the process of implementing the present invention, it is found that high-quality data sets are the key to improving the performance of learning-based wireless sensing. However, in the field of wireless sensing in the related art, there is still a lack of large-scale, high-quality public data resources, and data collection often relies on developers to collect and annotate by themselves, resulting in high development costs for wireless sensing models. In addition, since the wireless sensing model needs to be deployed together with the sensing device in the IoT device, the parameters of the wireless sensing model are more susceptible to leakage risks, thereby causing potential economic losses to the developer. Model watermarking technology is a common method for protecting deep models, which can prevent unauthorized copying and abuse. By adding watermarks to the parameters of the wireless sensing model, that is, by embedding specific identifiers or information, the unique identification of the wireless sensing model can be achieved, and the watermark is usually an invisible small parameter change, which has little effect on the overall performance of the wireless sensing model, but is sufficient to identify the source of the wireless sensing model. When the wireless sensing model is copied or abused, the watermark can trace its source, thereby providing developers with a powerful means of copyright protection. However, the watermarking technology in the related art is difficult to effectively deal with forged watermark attacks. A forged watermark attack refers to an attacker embedding his own watermark on a wireless sensing model that has been embedded with a legitimate watermark, so that both the original owner of the wireless sensing model and the attacker can claim the copyright of the wireless sensing model. In addition, the wireless sensing model can also be protected by judicial means after the fact. However, once the parameters of the wireless sensing model are leaked, the developer needs to use the watermark as proof of ownership and initiate legal proceedings to pursue responsibility. However, this method is not efficient in actual scenarios. The judicial process and infringement tracing costs are high, and for developers, passive protection after the fact often cannot effectively avoid losses.

[0030] In view of this, an embodiment of the present invention provides an object authentication method of a wireless perception model, including: obtaining an intermediate frequency receiving signal, wherein the intermediate frequency receiving signal is obtained by mixing a transmitting signal and an echo signal; performing phase offset compensation on the intermediate frequency receiving signal based on a fixed weight to generate a wireless signal spectrum; inputting the wireless signal spectrum into a wireless perception model, and outputting a target wireless signal spectrum, wherein the wireless perception model includes multiple feature extraction layers for extracting features from the wireless signal spectrum, at least one feature extraction layer is embedded with a pass authentication module, and the pass authentication module is used to perform phase offset compensation on the current input based on a dynamic weight to authenticate the pass of the target object; determining the authentication result of the target object according to the target wireless signal spectrum and a preset spectrum.

[0031] Figure 1 An application scenario diagram of an object authentication method based on a wireless perception model according to an embodiment of the present invention is shown.

[0032] like Figure 1 As shown, the application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. 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.

[0033] 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).

[0034] 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.

[0035] 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.

[0036] It should be noted that the object authentication method of the wireless perception model provided in the embodiment of the present invention can generally be executed by the server 105. Accordingly, the object authentication device of the wireless perception model provided in the embodiment of the present invention can generally be set in the server 105. The object authentication method of the wireless perception model 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 object authentication device of the wireless perception model 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.

[0037] It should be understood that Figure 1 The number of the first terminal device, the second terminal device, the third terminal device, the network and the server in the embodiment is only for illustration. According to the implementation requirements, there may be any number of terminal devices, networks and servers.

[0038] Figure 2 A flow chart of an object authentication method of a wireless awareness model according to an embodiment of the present invention is shown.

[0039] like Figure 2 As shown, the object authentication method 200 of the wireless perception model of this embodiment includes operations S210 to S240.

[0040] In operation S210, an intermediate frequency received signal is acquired.

[0041] In operation S220, fixed weight-based phase offset compensation is performed on the intermediate frequency received signal to generate a wireless signal spectrum.

[0042] In operation S230, the wireless signal spectrum is input into the wireless perception model, and the target wireless signal spectrum is output.

[0043] In operation S240, an authentication result of the target object is determined based on the target wireless signal spectrum and the preset spectrum.

[0044] According to an embodiment of the present invention, the intermediate frequency receiving signal may be obtained by mixing the transmitting signal and the echo signal. The transmitting signal is a signal transmitted by a wireless access point using an array antenna. The transmitting signal may be a wireless signal. The echo signal is a signal that is reflected back to the array antenna when the transmitting signal encounters a target object. A wireless access point may refer to a device that allows a wireless device to connect to a wired network. For example, the wireless device may be a mobile phone, a laptop computer, a tablet computer, etc. An array antenna may represent a collection of multiple antenna units. By controlling the phase and amplitude of the antenna unit, directional radiation of the beam may be achieved. Each antenna unit may be considered as an antenna, and the array antenna may be composed of multiple antennas.

[0045] According to an embodiment of the present invention, the intermediate frequency receiving signal may include K frequency points. By mixing the transmission signal and the echo signal, the intermediate frequency receiving signal may be obtained, as shown in the following formula (1).

[0046] (1)

[0047] in, It means that the mth antenna receives the intermediate frequency signal at the kth frequency point. represents the complex reflection coefficient of the target object located at (x, y) to the transmitted signal, j represents the imaginary unit, Indicates the round-trip transmission distance of the transmitted signal. Indicates the wavelength of the intermediate frequency received signal at the kth frequency point.

[0048] According to an embodiment of the present invention, by coherently superimposing intermediate frequency receiving signals of different antennas of the array antenna and different frequencies, phase offset compensation based on fixed weights can be performed on the intermediate frequency receiving signal, thereby generating a wireless signal spectrum.

[0049] According to an embodiment of the present invention, a fixed weight can be determined based on the echo signal and the transmitted signal. When the fixed weight is determined, a wireless signal spectrum can be generated by performing phase offset compensation on the intermediate frequency received signal based on the fixed weight.

[0050] According to an embodiment of the present invention, a wireless sensing model may represent a model that senses without interfering with the normal activities of a target object. For example, when the wireless sensing model senses the contour of a human body, the target object may be a human body. The wireless sensing model may include multiple feature extraction layers for extracting features from a wireless signal spectrum, at least one feature extraction layer being embedded with a pass authentication module, the pass authentication module being used to perform phase offset compensation based on dynamic weights on the current input to authenticate the pass of the target object.

[0051] For example, the wireless sensing model may be a neural network model. The feature extraction layer may be a convolutional layer, a fully connected layer, a recurrent neural network layer, an activation function, a normalization layer, etc. In addition, these feature extraction layers may be combined to form a wireless sensing model according to task requirements.

[0052] According to an embodiment of the present invention, by inputting the wireless signal spectrum into the wireless sensing model, the pass authentication module performs phase offset compensation based on dynamic weights on the current input to authenticate the pass of the target object, thereby outputting the target wireless signal spectrum. The pass can be a digital matrix, such as a QR code, a digital picture, etc. The target object can include multiple, and the pass of each target object is different.

[0053] According to an embodiment of the present invention, the preset spectrum can be set based on experience or can be a wireless signal spectrum. By comparing the quality of the target wireless signal spectrum and the preset spectrum, the authentication result of the target object can be determined, and the authentication result of the target object can indicate whether the target object has the authority to use the wireless perception model.

[0054] According to an embodiment of the present invention, a wireless signal spectrum can be generated by performing fixed-weight phase offset compensation on the acquired intermediate frequency received signal, and then the wireless signal spectrum is processed by the wireless perception model to obtain a target wireless signal spectrum. By comparing the target wireless signal spectrum with a preset spectrum, an authentication result of the target object is obtained. Since at least one of the multiple feature extraction layers in the wireless perception model is embedded with a pass authentication module, the pass authentication module performs dynamic weight-based phase offset compensation on the current input, and the pass of the target object can be authenticated. When the quality of the target wireless signal spectrum is higher than the preset spectrum, the target object is an authorized object. When the quality of the target wireless signal spectrum is lower than the preset spectrum, the target object is an unauthorized object. Therefore, when the target object is an authorized object, the quality of the target wireless signal spectrum is improved, thereby improving the perception accuracy of the wireless perception model. Since the target object can be authenticated by the quality of the output target wireless signal spectrum, active protection of the wireless perception model is achieved.

[0055] According to an embodiment of the present invention, a wireless signal spectrum is input into a wireless perception model, and a target wireless signal spectrum is output, including: determining a dynamic weight for a target feature extraction layer based on a preset pass; performing phase offset compensation on the current input based on the dynamic weight to generate an output of the target feature extraction layer.

[0056] According to an embodiment of the present invention, the target feature extraction layer may be a feature extraction layer in which a pass authentication module is embedded in the wireless perception model, and the target feature extraction layer may be one or more of the multiple feature extraction layers. For example, when the multiple feature extraction layers of the wireless perception model are respectively a convolutional layer, a fully connected layer, a recurrent neural network layer, an activation function, and a normalization layer, the target feature extraction layer may be a convolutional layer.

[0057] According to an embodiment of the present invention, the preset pass may include multiple ones, for example, the preset pass may be 2. For the target feature extraction layer, based on the preset pass, determining the dynamic weight includes: inputting the preset pass into the target feature extraction layer, outputting the first target feature; and determining the dynamic weight based on the first target feature.

[0058] According to an embodiment of the present invention, the first target feature may represent a feature obtained by the target feature extraction layer by extracting features from a preset pass. By inputting the preset pass into the target feature extraction layer, the target feature extraction layer extracts features from the preset pass to obtain the first target feature. Based on the first target feature, a dynamic weight may be determined, and the dynamic weight has periodicity.

[0059] According to an embodiment of the present invention, the input of the first feature extraction layer is the wireless signal spectrogram, the output of the last feature extraction layer is the target wireless signal spectrogram, and the output of the previous layer of the target feature extraction layer is the current input of the target feature extraction layer.

[0060] According to an embodiment of the present invention, phase offset compensation may be performed on the current input based on a dynamic weight, thereby generating an output of a target feature extraction layer.

[0061] According to an embodiment of the present invention, a preset pass can be input into the target feature extraction layer to obtain a first target feature, and then based on the first target feature, a dynamic weight can be determined. Based on the dynamic weight, the interference of the intermediate frequency receiving signal can be better suppressed, the super-resolution of the signal can be completed, and a higher quality target wireless signal spectrum can be obtained.

[0062] According to an embodiment of the present invention, based on dynamic weights, phase offset compensation is performed on the current input to generate the output of the target feature extraction layer, including: using the target feature extraction layer to extract features of the current input to obtain a second target feature; performing an inverse Fourier transform on the second target feature to obtain an original signal; and generating the output of the target feature extraction layer based on the original signal and the dynamic weights.

[0063] According to an embodiment of the present invention, the target feature extraction layer may be used to extract features from the current input to obtain a second target feature. The feature extraction process for the current input may be to perform an Inverse Fast Fourier Transform (IFFT) on the current input.

[0064] According to an embodiment of the present invention, the original signal may be obtained by performing a Fast Fourier Transform (FFT) on the second target feature.

[0065] Figure 3 The figure shows an architecture diagram of a pass authentication module according to an embodiment of the present invention.

[0066] like Figure 3 As shown, when the target feature extraction layer is a convolutional layer, the current input and the preset pass can be input into the convolutional layer, and the convolutional layer performs the current input By performing feature extraction, the second target feature can be obtained , and then for the second target feature Performing an inverse Fourier transform (IFFT) can get the original signal In addition, the convolution layer can also extract the features of preset pass a and preset pass p to obtain the first target feature corresponding to preset pass a. and the first target feature corresponding to the preset pass p , according to the first target feature corresponding to the preset pass a and the first target feature corresponding to the preset pass p , generating dynamic weights , for dynamic weights and the original signal After the product processing, Fourier transform (FFT) is performed to obtain the output of the target feature extraction layer.

[0067] According to an embodiment of the present invention, the output of the target feature extraction layer can be generated based on the original signal and the dynamic weight, including: determining the target feature of the echo signal based on the original signal and the dynamic weight; performing Fourier transform on the target feature of the echo signal to obtain the output of the target feature extraction layer.

[0068] According to an embodiment of the present invention, when the target feature extraction layer is a convolutional layer, Represents the convolution kernel, by passing the current input and convolution kernel By performing convolution processing, the second target feature can be obtained , as shown in formula (2). In the case where the preset passes include two, for example, the two preset passes are preset pass a and preset pass p, respectively, by combining preset pass a and convolution kernel By performing convolution processing, the first target feature corresponding to the preset pass a can be obtained , as shown in formula (3), by combining the preset pass p and the convolution kernel By performing convolution processing, the first target feature corresponding to the preset pass p can be obtained , as shown in formula (4).

[0069] (2)

[0070] (3)

[0071] (4)

[0072] According to an embodiment of the present invention, the original signal and the dynamic weight can be multiplied to obtain the target feature of the echo signal, as shown in the following formula (5). By performing Fourier transform on the target feature of the echo signal, the output of the target feature extraction layer can be obtained, as shown in the following formula (6).

[0073] (5)

[0074] (6)

[0075] Among them, s represents the target feature of the echo signal, represents the dynamic weight, represents the original signal, represents the inverse Fourier transform, represents the output of the target feature extraction layer, represents Fourier transform.

[0076] According to an embodiment of the present invention, by utilizing the target feature extraction layer to perform feature extraction on the current input, a second target feature can be obtained, and then the second target feature is subjected to Fourier transform to obtain the original signal, and by multiplying the original signal and the dynamic weight, the target feature of the echo signal can be obtained, and then the target feature of the echo signal is subjected to Fourier transform to obtain the output of the target feature extraction layer, thereby reducing the interference of the intermediate frequency receiving signal, realizing phase offset compensation for the echo signal based on dynamic weights, and improving the quality of the output of the target feature extraction layer.

[0077] According to an embodiment of the present invention, a phase offset compensation based on a fixed weight is performed on an intermediate frequency receiving signal to generate a wireless signal spectrum, including: determining a fixed weight according to the wavelength of the echo signal and the round-trip transmission distance of the transmitted signal; and based on the fixed weight, coherently superimposing the intermediate frequency receiving signal to generate a wireless signal spectrum.

[0078] According to an embodiment of the present invention, the round-trip transmission distance of the transmission signal may represent the transmission distance over which the transmission signal transmitted by the array antenna is reflected by the target object and then received by the array antenna.

[0079] According to an embodiment of the present invention, a fixed weight can be determined according to the wavelength of the echo signal and the round-trip transmission distance of the transmission signal, and then based on the fixed weight, the intermediate frequency receiving signal is coherently superimposed to generate a wireless signal spectrum, as shown in the following formula (7).

[0080] (7)

[0081] in, represents the wireless signal spectrum, K represents the number of frequency points, M represents the number of antennas in the array antenna, Represents a fixed weight.

[0082] According to an embodiment of the present invention, a fixed weight can be determined according to the wavelength of the echo signal and the round-trip transmission distance of the transmitted signal, and the intermediate frequency receiving signal can be coherently superimposed based on the fixed weight, so that the intermediate frequency receiving signal from the target position is enhanced and the intermediate frequency receiving signal at other positions is suppressed, so that the intermediate frequency receiving signal from the target position is separated, thereby generating a wireless signal spectrum.

[0083] According to an embodiment of the present invention, an authentication result of a target object is determined based on a target wireless signal spectrum and a preset spectrum, including: when the target wireless signal spectrum is a first target wireless signal spectrum, determining a first authentication result of the target object; when the target wireless signal spectrum is a second target wireless signal spectrum, determining a second authentication result of the target object.

[0084] According to an embodiment of the present invention, the preset spectrum can be a spectrum set according to demand or a wireless signal spectrum. The quality of the first target wireless signal spectrum is higher than the quality of the preset spectrum, and the first authentication result can indicate that the target object is an authorized object.

[0085] According to an embodiment of the present invention, when the target wireless signal spectrum is a first target wireless signal spectrum, the qualities of the first target wireless signal spectrum and a preset spectrum can be compared, so that a first authentication result of the target object can be determined.

[0086] According to an embodiment of the present invention, the quality of the second target wireless signal spectrum is lower than the quality of the preset spectrum, and the second authentication result can characterize the target object as an unauthorized object. In the case where the target wireless signal spectrum is the second target wireless signal spectrum, the quality of the second target wireless signal spectrum and the preset spectrum can be compared, so that the second authentication result of the target object can be determined.

[0087] According to an embodiment of the present invention, by comparing the first target wireless signal spectrum, the second target wireless signal spectrum and the preset spectrum, the first authentication result of the target object or the second authentication result of the target object can be obtained, thereby determining whether the target object is an authorized object, thereby realizing active protection of the wireless perception model and improving the protection flexibility of the wireless perception model.

[0088] Figure 4A A wireless signal spectrum diagram according to an embodiment of the present invention is shown. Figure 4B A first target wireless signal spectrum diagram according to an embodiment of the present invention is shown. Figure 4C A second target wireless signal spectrum diagram according to an embodiment of the present invention is shown.

[0089] Combination Figures 4A to 4C ,like Figure 4B The quality of the first target wireless signal spectrum shown is higher than that of Figure 4A The wireless signal spectrum shown in Figure 4C The quality of the second target wireless signal spectrum shown is lower than that of Figure 4A The wireless signal spectrum shown, that is, the target object corresponding to the first target wireless signal spectrum is an authorized object, and the target object corresponding to the second target wireless signal spectrum is an unauthorized object.

[0090] Figure 5 A flow chart of a watermark embedding method of a wireless perception model according to an embodiment of the present invention is shown.

[0091] like Figure 5 As shown, the watermark embedding method 500 of the wireless perception model of this embodiment includes operations S510 to S560.

[0092] In operation S510, a first target feature is determined according to a preset pass and a target feature extraction layer.

[0093] In operation S520, a dynamic weight is determined according to the first target feature.

[0094] In operation S530, a pass corresponding to the target object is generated according to the dynamic weight.

[0095] In operation S540 , the first target features are averaged to obtain a mean value of the first target features.

[0096] In operation S550, a character string watermark is determined according to the mean value of the first target feature.

[0097] In operation S560, the character string watermark is adjusted according to the preset watermark to obtain a target character string watermark, and the target character string watermark is embedded in the wireless perception model.

[0098] According to an embodiment of the present invention, the target feature extraction layer may be a feature extraction layer in which a pass authentication module is embedded in a wireless sensing model, and the target extraction layer may be one layer or multiple layers. For example, when the wireless sensing model includes a convolutional layer, a fully connected layer, a recurrent neural network layer, an activation function, and a normalization layer, and the pass authentication module is embedded in the convolutional layer, the convolutional layer is the target feature extraction layer.

[0099] According to an embodiment of the present invention, by inputting a preset pass into a target feature extraction layer, a first target feature can be obtained, and then a dynamic weight can be determined. Since the dynamic weight is periodic, a pass corresponding to the target object can be determined.

[0100] According to an embodiment of the present invention, by averaging the first target feature, the mean of the first target feature can be obtained. When the preset pass includes the preset pass a and the preset pass p, the means of the two first target features can be obtained, as shown in the following formulas (8) and (9), respectively. The mean of the first target feature is given a positive or negative sign to form a string watermark. The preset watermark is set according to demand. Based on the preset watermark, the parameters of the wireless sensing model can be adjusted to adjust the generated string watermark to obtain the target string watermark. The embedding of the target string watermark is achieved by adding a watermark embedding loss function in the subtask module, as shown in the following formula (10).

[0101] (8)

[0102] (9)

[0103] (10)

[0104] Among them, R represents the real matrix, W represents the width of the real matrix, H represents the height of the real matrix, and C represents the number of channels of the real matrix. Indicates the first target feature corresponding to the preset pass a The mean of Represents the first target feature corresponding to the preset pass p The mean of represents the index of the real matrix with width W, represents the index of the real matrix in height, Represents the first target feature In the values ​​of width index and height index, Represents the first target feature In the values ​​of width index and height index, represents the watermark embedding loss function, Represents the control parameter, the value of the control parameter can be 0.1, , B represents the preset watermark, , P represents the mean of the first target feature, represents the mean of the i-th first target feature, Represents the i-th preset watermark.

[0105] Figure 6 A diagram showing an example of generating a target character string watermark according to an embodiment of the present invention is shown.

[0106] like Figure 6 As shown, when the target feature extraction layer is a convolution layer, the preset pass a and the preset pass p are input into the convolution layer, and the first target feature corresponding to the preset pass a can be obtained. and the first target feature corresponding to the preset pass p , respectively for the first target feature corresponding to the preset pass a and the first target feature corresponding to the preset pass p By averaging, we can get the first target feature corresponding to the preset pass a. The mean and the first target feature corresponding to the preset pass p The mean , for the first target feature corresponding to the preset pass a The mean and the first target feature corresponding to the preset pass p The mean Taking the positive and negative signs, we can get the target string watermark 1, -1, 1, ..., 1, 1, -1, -1, ..., 1. Since the target string watermark is composed of the first target feature corresponding to the preset pass a The mean and the first target feature corresponding to the preset pass p The mean The length of the target string is twice the number of channels of the real matrix.

[0107] According to an embodiment of the present invention, the first target feature corresponding to the preset pass a The mean It can also be called the mean of the amplitude, the first target feature corresponding to the preset pass p The mean It can also be called the mean of the phase.

[0108] According to an embodiment of the present invention, the parameters of the wireless perception model are adjusted based on a preset watermark, thereby adjusting the generated string watermark to obtain a target string watermark, and then based on the wireless perception model after parameter adjustment, a target pass corresponding to the target object can be generated.

[0109] According to an embodiment of the present invention, different target passes can be assigned to different target objects based on the periodicity of dynamic weights, and the periodicity of dynamic weights satisfies the following formula (11). By utilizing the periodicity of dynamic weights, a target pass can be generated without changing the value of dynamic weights. The target pass satisfies the following formula (12). The target pass and the preset pass have the same function. A pass can be randomly initialized for the target object, and then the parameters of the wireless perception model can be adjusted using the gradient descent algorithm, that is, the operation of the following formula (13) is repeatedly performed until the value of the pass does not change, and the pass can be determined to be a target pass, that is, the target pass can be generated by the wireless perception model after adjusting the parameters.

[0110] (11)

[0111] (12)

[0112] (13)

[0113] Where k represents a positive integer matrix, represents the learning rate, A target pass representing a target object.

[0114] Figure 7 An example diagram of generating a target pass according to an embodiment of the present invention is shown.

[0115] like Figure 7 As shown, when the target feature extraction layer is a convolution layer, the preset pass a and the preset pass p are input into the convolution layer, and the first target feature corresponding to the preset pass a can be obtained. and the first target feature corresponding to the preset pass p , and then we can get the dynamic weight , since the dynamic weight is periodic, the dynamic weight satisfies , using the gradient descent algorithm, When it approaches 0, you can get the target pass .

[0116] According to an embodiment of the present invention, by adjusting the parameters of the target feature extraction layer in the wireless perception model, the string watermark can be adjusted to obtain the target string watermark, and the target string watermark is embedded in the wireless perception model based on the watermark embedding loss function of the above formula (10), so as to generate a target pass corresponding to the target object based on the wireless perception model embedded with the target string watermark.

[0117] According to an embodiment of the present invention, by inputting a preset pass into a target feature extraction layer, the first target feature can be determined, thereby obtaining a dynamic weight, and then generating a pass corresponding to the target object, and averaging the first target feature to obtain the mean of the first target feature. The mean of the first target feature is taken to form a string watermark, and the string watermark is adjusted based on the preset watermark to adjust the parameters of the wireless perception model to obtain the target string, and the target string watermark is embedded in the wireless perception model, so that the wireless perception model embedded with the target string watermark generates a target pass, thereby making the target object with the target pass an authorized object, thereby realizing active protection of the wireless perception model.

[0118] Figure 8 An example diagram of the perception process of the wireless perception model according to an embodiment of the present invention is shown.

[0119] like Figure 8 As shown, by preprocessing the echo signal, a wireless sensing signal spectrum can be obtained, wherein the preprocessing of the echo signal can be to mix the echo signal with the transmission signal, the purpose of which is to obtain an intermediate frequency receiving signal, and the intermediate frequency receiving signal is compensated for the phase offset based on a fixed weight to obtain a wireless signal spectrum. Then, the wireless sensing signal spectrum is input into the wireless sensing model, and the wireless sensing model generates different dynamic weights for different subtask modules, such as the 3D posture estimation subtask module and the human body contour generation subtask module, and performs phase offset compensation on the wireless signal spectrum based on different dynamic weights, thereby generating a target wireless signal spectrum, and then determining whether the target object using the wireless sensing model is an authorized object.

[0120] Fig. 9A The figure shows an architecture diagram of a wireless awareness model according to an embodiment of the present invention. Fig. 9B The standard modules of the wireless sensing model according to the embodiment of the present invention are shown. Fig. 9C The protection module of the wireless awareness model according to the embodiment of the present invention is shown.

[0121] In one embodiment, the subtask module of the wireless sensing model can be 3D posture estimation, which is a structure prediction task, using ConvNext as the backbone network, such as Fig. 9AAs shown, the wireless signal may be an intermediate frequency receiving signal, ConvNext may include 12 basic modules, and the standard ConvNext uses 12 standard modules, such as Fig. 9B As shown, the first 9 modules of the protected ConvNext adopt standard modules, and the last 3 adopt protection modules embedded with multi-user target pass authentication, as shown Fig. 9C As shown. The backbone network ConvNext is used to extract the features of the intermediate frequency received signal, and then the extracted features are input into the 3D attitude estimation subtask module to complete the attitude estimation. The performance measurement indicator is the average joint point error. The data set contains 47,304 samples in the training set and 22,192 samples in the test set. The wireless perception model is trained on the training set, and the performance of the wireless perception model is evaluated on the test set. The performance results show that the error of the 3D attitude estimation model using the standard ConvNext as the backbone network is 90.89 mm, and the error of the 3D attitude estimation model using the protected ConvNext as the backbone network is 87.46 mm. The pass authentication module will not only not affect the performance of the wireless perception model, but can improve the performance of the wireless perception model because the pass authentication module can provide super-resolution features. When the target object is an unauthorized object, that is, when an incorrect pass is given, the error of the 3D pose estimation model is 983.45mm, which shows the effectiveness of object authentication. The wireless perception model can work normally only when the target object is an authorized object, that is, when the correct pass is given. When the pass is wrong, the performance of the wireless perception model decreases significantly. When the target pass is input, the error of the 3D pose estimation model is 87.18mm, and the performance is consistent with the preset pass, and there is a slight performance improvement, that is, the target pass generation process of multiple users is lossless. When the 3D pose estimation model is attacked by a forged pass, that is, assuming that the attacker knows the parameters of the 3D pose estimation model and can obtain the complete training data, the parameters of the 3D pose estimation model are frozen, the pass is randomly initialized, and then the complete data set is used and the pass is updated using the gradient to obtain a forged pass. The error of the 3D pose estimation model is 147.53mm. Therefore, the pass authentication module can well resist reverse engineering or forged pass attacks. A pruning attack is performed on the 3D pose estimation model, and then the matching degree of the watermark is tested. Assuming that the attacker obtains the complete parameters of the model, a new model is obtained by trimming the redundant parameters of the model. Under different pruning rates, the pose error and watermark accuracy are shown in Table 1.

[0122] Table 1. Posture error and watermark accuracy under different pruning rates

[0123]

[0124] As shown in Table 1, even when the pruning rate is 90%, the watermark accuracy is still as high as 85.10%.

[0125] In one embodiment, a fine-tuning attack is used on the wireless perception model, that is, the backbone network parameters of the wireless perception model whose subtask is changed to human body contour generation are used to initialize the backbone network of the 3D posture estimation model, and then the 3D posture estimation data set is used for further fine-tuning. The watermark accuracy is 100%, and the error of the 3D posture estimation model is 99.59mm. Even if the model is used for other subtasks, the success rate of the embedded watermark being detected is still very high. Therefore, the 3D posture estimation model can be well protected.

[0126] In one embodiment, the subtask module of the wireless sensing model can be human body contour generation, which is a dense prediction task, using ConvNext as the backbone network, such as Fig. 9A As shown, ConvNext contains 12 basic modules, and the standard ConvNext uses 12 standard modules, such as Fig. 9B As shown, the first 9 modules of the protected ConvNext adopt standard modules, and the last 3 adopt protection modules with multi-user pass authentication, as shown Fig. 9CAs shown. The backbone network ConvNext is used to extract the features of the intermediate frequency received signal, which is then input into the decoder module to generate the human body contour. The performance measurement indicator is the average intersection over union (IoU). The dataset contains 17,520 samples in the training set and 4,675 samples in the test set. The wireless perception model is trained on the training set, and the performance of the wireless perception model is evaluated on the test set. The performance results show that the performance of the human body contour generation model using the standard ConvNext as the backbone network is 0.683 IoU, and the error of the human body contour generation model using the protected ConvNext as the backbone network is 0.684 IoU. Therefore, the pass authentication module will not only not affect the performance of the human body contour generation model, but can improve the performance of the human body contour generation model, because the pass authentication module can provide super-resolution features. When the target object is an unauthorized object, that is, when an incorrect pass is given, the performance of the human contour generation model is 0.283 IoU. Therefore, it can be shown that the effectiveness of object authentication is that the human contour generation model can work normally only when the target object is an authorized object, and the performance of the human contour generation model drops significantly when the wrong pass is given. When the target pass is input, the performance of the human contour generation model is 0.685 IoU, which is consistent with the initial pass and has a slight performance improvement, that is, the target pass generation process of multiple users is lossless. By using a forged pass attack on the human contour generation model, that is, assuming that the attacker knows the details of the human contour generation model and can obtain complete training data, by freezing the parameters of the human contour generation model, randomly initializing the pass, and then using the complete data set and using the gradient to update the pass to obtain a forged pass, the performance of the human contour generation model is 0.536 IoU, which shows that the pass authentication module can well resist reverse engineering or forged pass attacks. A pruning attack is performed on the human body contour generation model, and then the matching degree of the watermark is tested. Assuming that the attacker obtains the complete parameters of the wireless sensing model, a new model is obtained by trimming the redundant parameters of the model. Under different pruning rates, the contour generation and watermark accuracy are shown in Table 2.

[0127] Table 2 Contour generation and watermark accuracy under different pruning rates

[0128]

[0129] As shown in Table 2, even when the pruning rate is 90%, the watermark accuracy is still as high as 99.50%.

[0130] In one embodiment, a fine-tuning attack is applied to the human body contour generation model, that is, the backbone network parameters of the subtask are changed to 3D posture estimation to initialize the backbone network of the human body contour generation model, and then the contour generation data set is used for further fine-tuning. The watermark accuracy is 100%, and the performance of the human body contour generation model is 0.642 IoU. Therefore, it shows that even if the human body contour generation model is used for other tasks, the success rate of the embedded watermark being detected is still very high.

[0131] Fig.10 The sensing process of the wireless sensing model according to another embodiment of the present invention is shown.

[0132] like Fig.10 As shown, by inputting the wireless sensing signal spectrum into the wireless sensing model, when an authorized target object uses the wireless sensing model, a target wireless signal spectrum with a quality higher than the preset spectrum can be generated, that is, the performance of the wireless sensing model is normal; when an unauthorized target object uses the wireless sensing model, a target wireless signal spectrum with a quality lower than the preset spectrum can be generated, that is, the performance of the wireless sensing model is degraded.

[0133] Based on the object authentication method of the wireless perception model, the present invention also provides an object authentication device of the wireless perception model. Fig.11 The device is described in detail.

[0134] Fig.11 A structural block diagram of an object authentication device based on a wireless perception model according to an embodiment of the present invention is shown.

[0135] like Fig.11 As shown, the object authentication device 1100 of the wireless perception model of this embodiment includes an acquisition module 1110 , a first compensation module 1120 , a second compensation module 1130 and a determination module 1140 .

[0136] The acquisition module 1110 is used to acquire an intermediate frequency received signal, wherein the intermediate frequency received signal is obtained by mixing the transmission signal and the echo signal. In one embodiment, the acquisition module 1110 can be used to perform the operation S210 described above, which will not be described in detail here.

[0137] The first compensation module 1120 is used to perform phase offset compensation based on a fixed weight on the intermediate frequency received signal to generate a wireless signal spectrum. In one embodiment, the first compensation module 1120 can be used to perform the operation S220 described above, which will not be described in detail here.

[0138] The second compensation module 1130 is used to input the wireless signal spectrum into the wireless perception model and output the target wireless signal spectrum, wherein the wireless perception model includes a plurality of feature extraction layers for extracting features from the wireless signal spectrum, and at least one feature extraction layer is embedded with a pass authentication module, which is used to perform phase offset compensation based on dynamic weights on the current input to authenticate the pass of the target object. In one embodiment, the second compensation module 1130 can be used to perform the operation S230 described above, which will not be described in detail here.

[0139] The determination module 1140 is used to determine the authentication result of the target object according to the target wireless signal spectrum and the preset spectrum. In one embodiment, the determination module 1140 can be used to perform the operation S240 described above, which will not be repeated here.

[0140] According to an embodiment of the present invention, the second compensation module 1130 includes: a first compensation submodule and a second compensation submodule.

[0141] The first compensation submodule is used to determine a dynamic weight for a target feature extraction layer based on a preset pass, wherein the target feature extraction layer is a feature extraction layer in which a pass authentication module is embedded in a wireless perception model.

[0142] The second compensation submodule is used to perform phase offset compensation on the current input based on the dynamic weight to generate the output of the target feature extraction layer, wherein the input of the first feature extraction layer is the wireless signal spectrum, the output of the last feature extraction layer is the target wireless signal spectrum, and the output of the previous layer of the target feature extraction layer is the current input of the target feature extraction layer.

[0143] According to an embodiment of the present invention, a first compensation submodule includes: a first compensation unit and a second compensation unit.

[0144] The first compensation unit is used to input the preset pass into the target feature extraction layer and output the first target feature, wherein the first target feature represents the feature obtained by the target feature extraction layer by extracting the preset pass.

[0145] The second compensation unit is used to determine the dynamic weight based on the first target feature.

[0146] According to an embodiment of the present invention, the second compensation submodule includes: a third compensation unit, a fourth compensation unit and a fifth compensation unit.

[0147] The third compensation unit is used to use the target feature extraction layer to extract features from the current input to obtain a second target feature.

[0148] The fourth compensation unit is used to perform inverse Fourier transform on the second target feature to obtain an original signal.

[0149] The fifth compensation unit is used to generate the output of the target feature extraction layer according to the original signal and the dynamic weight.

[0150] According to an embodiment of the present invention, a fifth compensation unit includes: a first compensation subunit and a second compensation subunit.

[0151] The first compensation subunit is used to determine the target feature of the echo signal according to the original signal and the dynamic weight.

[0152] The second compensation subunit is used to perform Fourier transform on the target feature of the echo signal to obtain the output of the target feature extraction layer.

[0153] According to an embodiment of the present invention, the first compensation module 1120 includes: a third compensation submodule and a fourth compensation submodule.

[0154] The third compensation submodule is used to determine a fixed weight according to the wavelength of the echo signal and the round-trip transmission distance of the transmitted signal, wherein the round-trip transmission distance of the transmitted signal represents the transmission distance of the transmitted signal emitted by the array antenna after being reflected by the target object and then received by the array antenna.

[0155] The fourth compensation submodule is used to perform coherent superposition on the intermediate frequency received signal based on a fixed weight to generate a wireless signal spectrum.

[0156] According to an embodiment of the present invention, the determination module 1140 includes: a first determination submodule and a second determination submodule.

[0157] The first determination submodule is used to determine a first authentication result of the target object when the target wireless signal spectrum is a first target wireless signal spectrum, wherein the quality of the first target wireless signal spectrum is higher than the quality of a preset spectrum, and the first authentication result characterizes that the target object is an authorized object.

[0158] The second determination submodule is used to determine the second authentication result of the target object when the target wireless signal spectrum is a second target wireless signal spectrum, wherein the quality of the second target wireless signal spectrum is lower than the quality of the preset spectrum, and the second authentication result characterizes that the target object is an unauthorized object.

[0159] According to an embodiment of the present invention, any multiple modules among the acquisition module 1110, the first compensation module 1120, the second compensation module 1130 and the determination module 1140 can be combined into one module for implementation, 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 acquisition module 1110, the first compensation module 1120, the second compensation module 1130 and the determination module 1140 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 a proper combination of any of them. Alternatively, at least one of the acquisition module 1110 , the first compensation module 1120 , the second compensation module 1130 , and the determination module 1140 may be at least partially implemented as a computer program module, and when the computer program module is executed, a corresponding function may be performed.

[0160] Fig.12 A block diagram of an electronic device suitable for implementing an object authentication method of a wireless perception model according to an embodiment of the present invention is shown.

[0161] like Fig.12 As shown, the electronic device 1200 according to an embodiment of the present invention includes a processor 1201, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 1202 or the program loaded from the storage part 1208 to the RAM (Random Access Memory). The processor 1201 may include, for example, a general-purpose microprocessor (such as a CPU), an instruction set processor and / or a related chipset and / or a special-purpose microprocessor (for example, an application-specific integrated circuit (ASIC)), etc. The processor 1201 may also include an onboard memory for caching purposes. The processor 1201 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.

[0162] In RAM 1203, various programs and data required for the operation of electronic device 1200 are stored. Processor 1201, ROM 1202 and RAM 1203 are connected to each other via bus 1204. Processor 1201 performs various operations of the method flow according to the embodiment of the present invention by executing the program in ROM 1202 and / or RAM 1203. It should be noted that the program can also be stored in one or more memories other than ROM 1202 and RAM 1203. Processor 1201 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.

[0163] According to an embodiment of the present invention, the electronic device 1200 may further include an input / output (I / O) interface 1205, which is also connected to the bus 1204. The electronic device 1200 may further include one or more of the following components connected to the input / output (I / O) interface 1205: an input portion 1206 including a keyboard, a mouse, etc.; an output portion 1207 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage portion 1208 including a hard disk, etc.; and a communication portion 1209 including a network interface card such as a LAN card, a modem, etc. The communication portion 1209 performs communication processing via a network such as the Internet. A drive 1210 is also connected to the input / output (I / O) interface 1205 as needed. A removable medium 1211, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1210 as needed, so that a computer program read therefrom is installed into the storage portion 1208 as needed.

[0164] 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.

[0165] 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 RAM (Random Access Memory), a ROM (Read Only Memory), 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 1202 and / or RAM 1203 described above and / or one or more memories other than ROM 1202 and RAM 1203.

[0166] The embodiment of the present invention also includes a computer program product, which includes a computer program, and the computer program contains program code for executing the method shown in the flowchart. When the computer program product is run in a computer system, the program code is used to enable the computer system to implement the object authentication method of the wireless perception model provided by the embodiment of the present invention.

[0167] The computer program executes the above functions defined in the system / device of the embodiment of the present invention when it is executed by the processor 1201. According to the embodiment of the present invention, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0168] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices, magnetic storage devices, etc. In another embodiment, the computer program may also be transmitted and distributed in the form of signals on a network medium, and downloaded and installed through the communication part 1209, and / or installed from the removable medium 1211. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0169] In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 1209, and / or installed from the removable medium 1211. When the computer program is executed by the processor 1201, the above functions defined in the system of the embodiment of the present invention are performed. According to the embodiment of the present invention, the system, device, means, module, unit, etc. described above can be implemented by a computer program module.

[0170] According to an embodiment of the present invention, the program code for executing the computer program provided by the embodiment of the present invention can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level process and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, Java, C++, python, "C" language or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on the remote computing device, or entirely on the remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0171] 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.

[0172] 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.

[0173] 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 sensing model object authentication method, characterized in that: The method comprises: Acquire an intermediate frequency receiving signal, wherein the intermediate frequency receiving signal is obtained by mixing a transmitting signal and an echo signal; Performing phase offset compensation based on a fixed weight on the intermediate frequency received signal to generate a wireless signal spectrum; Input the wireless signal spectrum to a wireless sensing model, and output a target wireless signal spectrum, wherein the wireless sensing model includes a plurality of feature extraction layers for extracting features from the wireless signal spectrum, and at least one of the feature extraction layers is embedded with a pass authentication module, and the pass authentication module is used to perform phase offset compensation based on dynamic weights on the current input to authenticate the pass of the target object; An authentication result of the target object is determined based on the target wireless signal spectrum and a preset spectrum.

2. The method according to claim 1, characterized in that The step of inputting the wireless signal spectrum graph into a wireless perception model and outputting a target wireless signal spectrum graph comprises: For a target feature extraction layer, the dynamic weight is determined based on a preset pass, wherein the target feature extraction layer is a feature extraction layer in which the pass authentication module is embedded in the wireless sensing model; Based on the dynamic weight, the current input is phase offset compensated to generate the output of the target feature extraction layer, wherein the input of the first feature extraction layer is the wireless signal spectrum, the output of the last feature extraction layer is the target wireless signal spectrum, and the output of the previous layer of the target feature extraction layer is the current input of the target feature extraction layer.

3. The method according to claim 2, characterized in that The step of determining the dynamic weight for the target feature extraction layer based on a preset pass includes: Input the preset pass into the target feature extraction layer, and output a first target feature, wherein the first target feature represents a feature obtained by the target feature extraction layer by extracting features from the preset pass; Based on the first target feature, the dynamic weight is determined.

4. The method according to claim 2, characterized in that: The performing phase offset compensation on the current input based on the dynamic weight to generate the output of the target feature extraction layer includes: Using the target feature extraction layer to extract features from the current input, to obtain a second target feature; Performing an inverse Fourier transform on the second target feature to obtain an original signal; The output of the target feature extraction layer is generated according to the original signal and the dynamic weight.

5. The method according to claim 4, characterized in that Generating the output of the target feature extraction layer according to the original signal and the dynamic weight includes: Determining a target feature of the echo signal according to the original signal and the dynamic weight; Perform Fourier transform on the target feature of the echo signal to obtain the output of the target feature extraction layer.

6. The method according to claim 1, characterized in that The performing fixed weight-based phase offset compensation on the intermediate frequency received signal to generate a wireless signal spectrum diagram includes: Determine the fixed weight according to the wavelength of the echo signal and the round-trip transmission distance of the transmission signal, wherein the round-trip transmission distance of the transmission signal represents the transmission distance of the transmission signal transmitted by the array antenna after being reflected by the target object and then received by the array antenna; Based on the fixed weights, the intermediate frequency received signals are coherently superimposed to generate the wireless signal spectrum.

7. The method according to claim 1, characterized in that The step of determining the authentication result of the target object according to the target wireless signal spectrum and the preset spectrum includes: In a case where the target wireless signal spectrum is a first target wireless signal spectrum, determining a first authentication result of the target object, wherein the quality of the first target wireless signal spectrum is higher than the quality of the preset spectrum, and the first authentication result indicates that the target object is an authorized object; In a case where the target wireless signal spectrum is a second target wireless signal spectrum, a second authentication result of the target object is determined, wherein the quality of the second target wireless signal spectrum is lower than the quality of the preset spectrum, and the second authentication result characterizes that the target object is an unauthorized object.

8. A watermark embedding method for a wireless perception model, characterized in that: The method comprises: Determine a first target feature according to a preset pass and a target feature extraction layer, wherein the target feature extraction layer is a feature extraction layer in which a pass authentication module is embedded in a wireless sensing model; Determining a dynamic weight according to the first target feature; Generate a pass corresponding to the target object according to the dynamic weight; Averaging the first target features to obtain a mean value of the first target features; Determining a character string watermark according to the mean value of the first target feature; According to the preset watermark, the character string watermark is adjusted to obtain a target character string watermark, and the target character string watermark is embedded in the wireless perception model.

9. An object authentication device of a wireless perception model, characterized in that: The device comprises: An acquisition module, used for acquiring an intermediate frequency receiving signal, wherein the intermediate frequency receiving signal is obtained by mixing a transmission signal and an echo signal; A first compensation module, used for performing phase offset compensation on the intermediate frequency received signal based on a fixed weight to generate a wireless signal spectrum; a second compensation module, configured to input the wireless signal spectrum into a wireless sensing model and output a target wireless signal spectrum, wherein the wireless sensing model includes a plurality of feature extraction layers for extracting features from the wireless signal spectrum, at least one of the feature extraction layers is embedded with a pass authentication module, and the pass authentication module is configured to perform phase offset compensation based on dynamic weights on the current input to authenticate the pass of the target object; A determination module is used to determine the authentication result of the target object based on the target wireless signal spectrum and a preset spectrum.

10. An electronic device, comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 7 or 8.

Images