MIMO communication system radio frequency fingerprint protection method based on structured perturbation

Abstract

The present application relates to the field of wireless communication physical layer security technology, especially to a MIMO communication system radio frequency fingerprint protection method based on structured disturbance. The method first acquires a communication signal to be sent and a pre-shared key, generates a disturbance parameter based on the pre-shared key and a data packet index; uses a first disturbance factor to independently perform weighting processing on a training sequence in the communication signal in the dimension of a transmission antenna, maintains the spatial orthogonal structure among multi-antenna signals; performs structured linkage disturbance processing on a data symbol in the communication signal, establishes a disturbance association between the data symbol and the training sequence; and sends the communication signal after the weighting processing and the structured linkage disturbance processing through a wireless channel. The present application can effectively hide the radio frequency fingerprint to prevent device tracking, maintain the spatial orthogonality of the MIMO preamble, and realize the recoverable authentication of the cooperative receiver and the transparent compatibility of the non-cooperative standard receiver.

Description

Technical Field

[0001] This invention relates to the field of wireless communication physical layer security technology, and in particular to a method for protecting the radio frequency fingerprint of a MIMO communication system based on structured perturbations. Background Technology

[0002] Radio Frequency Fingerprint (RFF) technology, as an emerging physical layer security authentication method, utilizes the unique, inherent, and uncopyable hardware defects caused by hardware tolerances (such as digital-to-analog converters, mixers, and power amplifiers) during the manufacturing process of wireless devices to provide a physical identity for the device. Compared with traditional security mechanisms that rely on upper-layer encryption algorithms, RFF authentication has advantages such as endogenousness, difficulty in forgery, and low computational overhead, making it particularly suitable for resource-constrained scenarios such as the Internet of Things (IoT) and the Internet of Vehicles (IoV). It serves as a powerful supplement to mainstream cryptography-based authentication mechanisms or software-based authentication methods. In recent years, researchers both domestically and internationally have conducted extensive research on the extraction, recognition, and authentication technologies of RF fingerprints. Especially with the rise of artificial intelligence, RF fingerprint technology has seen significant improvements in feature extraction, recognition accuracy, and real-time performance.

[0003] However, research on the security of RFF technology as a physical layer security authentication technology is insufficient. The uniqueness and stability of radio frequency fingerprints also pose serious privacy and security challenges. In broadcast channels, non-cooperative attackers can passively collect signals and extract fingerprint features, thereby enabling cross-location and cross-session device tracking without the user's knowledge. Because these features originate from hardware manufacturing tolerances, users cannot easily reset or revoke them through software updates, resulting in a persistent risk of physical layer privacy breaches.

[0004] To address this technical challenge, existing research often attempts to mask fingerprints by introducing noise, followed by signal compensation to eliminate the noise and recover the fingerprint. However, while achieving privacy protection, this method generally sacrifices the usability of RFF authentication, failing to meet the practical requirement of both concealment and authentication. Furthermore, existing technologies involve collaborating with partners to remove disturbances and recover the fingerprint for authentication at the receiving end via an agreement. This method overcomes the usability issue of RFF authentication, but it is mostly designed and verified in Single-Input Single-Output (SISO) systems, and still generally suffers from a series of problems related to protection strength and system compatibility. Summary of the Invention

[0005] The present invention aims to at least partially solve one of the technical problems in the related art.

[0006] Therefore, the first objective of this invention is to provide a radio frequency fingerprint protection method for a multi-antenna communication system, comprising the following steps:

[0007] S1, obtain the communication signal to be sent and the pre-shared key, and generate perturbation parameters based on the pre-shared key and the data packet index. The perturbation parameters include a first perturbation factor and a second perturbation factor. S2, using the first perturbation factor to perform independent weighting processing on the training sequence in the communication signal according to the transmit antenna dimension, so as to maintain the spatial orthogonal structure between the multi-antenna signals; S3, using the first perturbation factor and the second perturbation factor to perform structured linkage perturbation processing on the data symbols in the communication signal, and establishing a perturbation association between the data symbols and the training sequence; S4, the communication signal after the weighting process and the structured linkage disturbance process is transmitted through the wireless channel.

[0008] In one embodiment of the present invention, S1 includes: S11, Generate a data packet index for each data packet to be sent, and the data packet index and the pre-shared key are used together as seed input; S12, using the pre-shared key and the data packet index as input, call a cryptographic pseudo-random function to generate a binary bit stream; S13, the binary bit stream is converted into the first perturbation factor and the second perturbation factor using a mapping function. The first perturbation factor is used for weighted processing of the training sequence, and the second perturbation factor is used for structured linkage perturbation processing of the data symbols.

[0009] In one embodiment of the present invention, S13 further includes: S131, the binary bit stream is split into several bit groups, and each bit group is mapped to a perturbed alphabet index; S132, select amplitude parameters and phase parameters from the amplitude set and phase set based on the perturbation alphabet index, wherein the amplitude set is symmetrically designed around the nominal amplitude; S133, construct complex perturbation elements based on the amplitude parameter and the phase parameter, and assemble the complex perturbation elements into the first perturbation factor and the second perturbation factor, wherein the construction formula for a single perturbation element is: ; in, This represents the complex perturbation element. This indicates the amplitude parameter. This represents the phase parameter. It represents the imaginary unit.

[0010] In one embodiment of the present invention, S2 includes: S21, Construct a diagonal matrix based on the first perturbation factor, wherein the dimension of the diagonal matrix is ​​adapted to the number of transmit antennas, and wherein the training sequence is a preamble; S22, using the diagonal matrix to... The original preamble matrices on the subcarriers are left-multiplied and weighted to obtain the hidden preamble, where the ... The formula for calculating the hidden preamble on each subcarrier is: ; in, This represents the hidden preamble. Indicates that the first disturbance factor is in the first... Components on each subcarrier This represents the original preamble matrix. This indicates a diagonalization operation.

[0011] In one embodiment of the present invention, S3 includes: S31, reuse the first perturbation factor used for the weighting processing of the training sequence as the left multiplication factor of the data symbol; S32, the original data symbols are multiplied element by element using the second perturbation factor to obtain intermediate perturbation data symbols; S33, apply the left multiplication factor to the intermediate perturbation data symbol to obtain the hidden data symbol, where the first... The formula for calculating the hidden data symbols on each subcarrier is: ; in, This represents the hidden data symbol. Indicates that the second disturbance factor is in the first... Components on each subcarrier Represents the original data symbols. This indicates element-wise multiplication.

[0012] In one embodiment of the present invention, S4 includes: S41, the hidden preamble and the hidden data symbols are processed by digital domain modulation to obtain the time domain signal; S42, add a cyclic prefix to the time-domain signal to obtain a transmission frame that conforms to the orthogonal frequency division multiplexing standard; S43, the transmitted frame is sent to the wireless channel via the radio frequency front end for reception by cooperative receivers and non-cooperative standard receivers.

[0013] In one embodiment of the present invention, S43 further includes: S431, after the cooperative receiver receives the signal sent by the transmitter, it reconstructs the first perturbation factor and the second perturbation factor based on the corresponding key and data packet index; S432, perform corrected channel estimation based on the first perturbation factor, and perform channel equalization and perturbation removal on the received signal to obtain data symbols under noise-limited distortion only; S433 inputs the denoised corrected channel estimation result and the data symbols under noise-limited distortion only into the RF fingerprint extraction and matching module, and outputs the authentication result.

[0014] In one embodiment of the present invention, the second perturbation factor for the structured linkage perturbation processing of the data symbols includes: The second perturbation factor is constructed as a perturbation matrix that matches the dimension of the data symbols; The elements of the perturbation matrix are obtained by bit block mapping and are used to further obfuscate the data segment fingerprint; The element-wise multiplication improves unpredictability to unauthorized parties while maintaining parameter synchronization with cooperative receivers.

[0015] To achieve the above objectives, a second aspect of the present invention provides a radio frequency fingerprint protection system for a multi-antenna communication system, comprising: The disturbance parameter generation module is used to acquire the communication signal to be transmitted and the pre-shared key, and generate disturbance parameters based on the pre-shared key and the data packet index. The disturbance parameters include a first disturbance factor and a second disturbance factor. The weighting processing module is used to perform independent weighting processing on the training sequence in the communication signal according to the transmit antenna dimension using the first perturbation factor, so as to maintain the spatial orthogonal structure between the multi-antenna signals. The linkage perturbation processing module is used to perform structured linkage perturbation processing on the data symbols in the communication signal using the first perturbation factor and the second perturbation factor, and to establish a perturbation association between the data symbols and the training sequence; The transmitting module is used to transmit the communication signal after the weighting processing and the structured linkage perturbation processing through a wireless channel. The receiving module is used to receive signals sent by the transmitting module and perform radio frequency fingerprint extraction and matching after decoding.

[0016] To achieve the above objectives, a third aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in the first aspect.

[0017] The method, system, and storage medium of this invention can achieve reversible hiding and protection of radio frequency fingerprints, effectively suppressing malicious device tracking while supporting cooperative receivers to recover fingerprint authentication, and maintaining the spatial orthogonality of MIMO preambles to ensure communication compatibility of non-cooperative standard receivers.

[0018] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0019] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a structural diagram of a wireless system according to an embodiment of the present invention; Figure 2 This is a flowchart of a radio frequency fingerprint protection method for a MIMO communication system based on structured perturbation according to an embodiment of the present invention; Figure 3 This is a structural diagram of a radio frequency fingerprint protection system for a MIMO communication system based on structured perturbation, according to an embodiment of the present invention. Detailed Implementation

[0020] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0022] The following description, with reference to the accompanying drawings, describes a method and system for radio frequency fingerprint protection of a MIMO communication system based on structured perturbation, according to an embodiment of the present invention.

[0023] Example 1 Figure 2 This is a flowchart of a radio frequency fingerprint protection method for a multi-antenna communication system according to an embodiment of the present invention.

[0024] like Figure 1 As shown, in practical use, a wireless system generally includes the following entities: Transmitter: Shares secret key with cooperative receiver An RFF Masking module is embedded in the digital domain processing to implement a radio frequency fingerprint hiding security mechanism and communicate normally with non-cooperative standard receivers; Cooperative receiver: Shares a secret key with the transmitter By embedding an RFF Unmasking module in the digital domain processing, a reverse radio frequency fingerprint recovery mechanism is executed, and such receivers need to identify the transmitter's identity. Non-cooperative standard receivers: These receivers accept data packets from the transmitter, do not participate in security authentication protocols, and only perform standard communication demodulation. These receivers are commonly found in highly dynamic multicast systems, where frequent joining and leaving of receivers is a frequent occurrence. In some cases, malicious attackers may emerge in wireless systems. These attackers, often passive eavesdroppers, aim to perform RFF-based device identification and association across data packets. They can collect data packets from any number of transmitters and possess knowledge of modulation and coding schemes, as well as the common preamble structure. However, these malicious attackers do not possess the shared keys between the transmitters and cooperating receivers, thus preventing the implementation of RF fingerprinting security mechanisms.

[0025] The common MIMO-OFDM system model selected in this invention, its signal model, and the processing flow of the standard receiver are as follows: Assuming the transmitter has One antenna, one receiver Root antenna. OFDM systems have The subcarrier, at the subcarrier Subcarriers ( On the above, the transmitted signal matrix is ,in For preamble, For data symbols. This refers to the number of preamble symbols. Unless otherwise specified, it is generally set to the equivalent number of antennas by default. , It is a number of data symbols.

[0026] Signals pass through RF front-end impairments (such as I / Q imbalance, CFO, power amplifier nonlinearity), multipath channels It reaches the receiving end after being mixed with additive noise.

[0027] make This represents the equivalent end-to-end channel that aggregates the wireless channel and transmitter RF impairments (i.e., fingerprint characteristics). Standard receivers typically employ zero-forcing (ZF) equalization, and its processing steps are as follows: 1. Channel estimation: using a known preamble and the received preamble Channel estimation: ; 2. Data equalization: Utilizing the estimated channel... For the received data Perform recovery: .

[0028] like Figure 1 As shown, the radio frequency fingerprint protection method for multi-antenna communication systems includes the following steps: S1, obtain the communication signal to be sent and the pre-shared key, and generate perturbation parameters based on the pre-shared key and the data packet index. The perturbation parameters include a first perturbation factor and a second perturbation factor.

[0029] This step involves the generation and configuration of perturbation parameters during RF fingerprint protection. The transmitter first acquires the communication signal to be transmitted and a pre-agreed shared key with the cooperating receiver. Combining this with the packet index of the current data packet, dynamic perturbation parameters are generated using a cryptographic mechanism. These perturbation parameters are constructed as a combination of a first perturbation factor and a second perturbation factor, corresponding to the training sequence field and data symbol field in the communication signal, respectively, thus providing a basis for subsequent differentiated perturbation processing. The core of this process lies in using the pre-shared key to ensure the reproducibility of parameter generation, while using the packet index to ensure the independence of parameters across different transmission instances, preventing malicious statistical analysis of perturbation patterns. As one implementation method, a pseudo-random function based on the key can be invoked. and packet index Generate binary bit stream Then through the mapping function Convert to perturbation matrix ,in and They are applied as the first and second perturbation factors to different signal segments, respectively.

[0030] Specifically, step S1 includes: S11, Generate a data packet index for each data packet to be sent, and the data packet index and the pre-shared key are used together as seed input.

[0031] Specifically, the transmitting equipment and the cooperating receiving equipment share a secret key in advance. For each data packet to be sent, the transmitter generates a data packet index. (e.g., package number, nonce) This can be provided by a frame counter, session sequence number, timestamp hash, or a mutually negotiated random number. To prevent perturbation reuse, It is unique for each data packet.

[0032] S12, using the pre-shared key and the data packet index as input, call a cryptographic pseudo-random function to generate a binary bit stream.

[0033] Specifically, the transmitting end is based on Call a cryptographic pseudo-random function / pseudo-random generator for input. Generate binary bit stream . CTR-DRBG, HMAC-DRBG, or other similar implementations can be used; It is used as a nonce in seed construction to avoid cross-packet duplication.

[0034] S13, the binary bit stream is converted into the first perturbation factor and the second perturbation factor using a mapping function. The first perturbation factor is used for weighted processing of the training sequence, and the second perturbation factor is used for structured linkage perturbation processing of the data symbols.

[0035] Specifically, using mapping functions Bitstream Converted into two sets of perturbation matrices (i.e., the first perturbation factor) Second disturbance factor ): , ; , ; Specifically, in one embodiment, the bitstream to be embedded is... It is split into several bit blocks, each block is mapped to a perturbation alphabet index, and then selected from the amplitude set accordingly. With phase set Select amplitude parameter and phase parameters Construct complex perturbation elements This achieves joint quantization of "discrete amplitude and discrete phase". To improve perturbation representation while controlling overhead, and All are finite sets, and their orders are denoted as . , The alphabet size of a single perturbation element is Each element can hold Bit information. The system can select based on the desired level of privacy and the allowable error margin. , Size: Generally speaking, the larger the order, the richer the perturbation state and the stronger the confusion for the eavesdropper, but the requirements for implementation complexity and parameter synchronization also increase accordingly.

[0036] Regarding parameter selection, to ensure that the average transmit power remains constant and to avoid introducing statistically detectable offsets, the amplitude set... Typically, a symmetrical design is used around the nominal amplitude 1 (e.g.) ), and make the probability of each amplitude point symmetrical, thereby (or The phase set is established. An equally spaced phase point design is used to ensure that the disturbance does not introduce a systematic bias in a statistical sense. Furthermore, the mapping rule can be set to use each value symmetrically within a time window to maintain long-term statistical stability.

[0037] Based on the above perturbation elements, two types of perturbation structures are constructed as follows: Preamble perturbation vector Each component is obtained by mapping the corresponding group, and is used to construct a diagonal matrix. Matrix dimension and number of transmit antennas adaptation.

[0038] Data perturbation matrix Its elements are also obtained from group mapping, used to further obfuscate data symbols and increase unpredictability to unauthorized parties.

[0039] This step, through a key-driven dynamic parameter generation mechanism, ensures that the perturbation sequence is unpredictable for unauthorized parties while achieving parameter synchronization between authorized and unauthorized parties. This lays a secure foundation for the subsequent construction of a recoverable RF fingerprint hiding structure and effectively balances the strength of privacy protection with the reliability of authentication.

[0040] S2, using the first perturbation factor to perform independent weighting processing on the training sequence in the communication signal according to the transmit antenna dimension, so as to maintain the spatial orthogonal structure between the multi-antenna signals.

[0041] Specifically, step S2 includes: S21, Construct a diagonal matrix based on the first perturbation factor, wherein the dimension of the diagonal matrix is ​​adapted to the number of transmit antennas, and wherein the training sequence is a preamble; S22, using the diagonal matrix to... The original preamble matrix on each subcarrier is left-multiplied and weighted to obtain the hidden preamble.

[0042] Specifically, in MIMO systems, preambles are typically designed as orthogonal matrices to aid channel estimation. To hide the fingerprint without compromising the spatial orthogonality between multiple antennas in the MIMO system, this invention performs diagonalization perturbation on the preamble symbols. For the... There are subcarriers, and the original preamble matrix is ​​. Hidden preamble The calculation formula is: ; Among them, the standard MIMO preamble It is usually a unitary matrix. Apply a diagonal matrix. This is equivalent to performing independent complex weighting on the signal from each transmitting antenna.

[0043] Perform orthogonality verification on the preamble: .

[0044] The result is still a diagonal array, indicating that the signal inner product between different antennas is still zero. This ensures that the receiver will not experience inter-antenna interference during channel estimation.

[0045] By employing the aforementioned independent weighting processing based on the transmit antenna dimension, it is possible to strictly maintain the spatial orthogonality of multi-antenna signals while concealing device fingerprint characteristics. This ensures that non-cooperative standard receivers do not experience inter-antenna interference during channel estimation, guaranteeing the accuracy of channel estimation and compatibility with data demodulation. Consequently, privacy protection is achieved without affecting the normal performance of existing communication systems.

[0046] S3, using the first perturbation factor and the second perturbation factor to perform structured linkage perturbation processing on the data symbols in the communication signal, and establishing a perturbation association between the data symbols and the training sequence.

[0047] Specifically, in order to achieve "transparency" to the standard receiver, the hiding of data symbols needs to be mathematically linked to the hiding of the preamble. Step S3 includes: S31, reuse the first perturbation factor used for the weighting processing of the training sequence as the left multiplication factor of the data symbol; S32, the original data symbols are multiplied element by element using the second perturbation factor to obtain intermediate perturbation data symbols; S33, apply the left multiplication factor to the intermediate perturbation data symbol to obtain the hidden data symbol, where the first... The formula for calculating the hidden data symbols on each subcarrier is: ; in, This represents the hidden data symbol. Indicates that the second disturbance factor is in the first... Components on each subcarrier Represents the original data symbols. This represents element-wise multiplication (Hadamard product).

[0048] By establishing perturbation correlations between data symbols and training sequences, non-cooperative receivers can naturally absorb common perturbation components during channel estimation and equalization, thereby ensuring communication compatibility. Meanwhile, cooperative receivers can accurately remove perturbations based on the correlations to recover the radio frequency fingerprint, effectively improving the balance between privacy protection and authentication reliability.

[0049] S4, the communication signal after the weighting process and the structured linkage disturbance process is transmitted through the wireless channel.

[0050] S41, following the standard MIMO-OFDM process, performs digital domain modulation on the hidden preamble and hidden data symbols to obtain the time domain signal.

[0051] As one implementation method, digital domain modulation processing includes modulation methods such as inverse Fourier transform and adding a cyclic prefix.

[0052] S42, add a cyclic prefix to the time-domain signal to obtain a transmission frame that conforms to the orthogonal frequency division multiplexing standard; S43, the transmitted frame is sent to the wireless channel via the radio frequency front end for reception by cooperative receivers and non-cooperative standard receivers.

[0053] Specifically, the cooperating receiver possesses the key. The reconstruction steps are exactly the same as those at the transmitting end, referring to step one and using the key. and package index Reconstructed and In the first Cooperative receivers on each subcarrier utilize known preamble perturbations. Corrected channel estimation: ; Then the cooperative receiver receives the data symbols. Jointly perform channel equalization and disturbance removal: ; in This indicates element-wise division.

[0054] After the above steps, the cooperative receiver obtained It is a data symbol with perturbations removed and distortion limited only by noise, approximating the original signal, with negligible degradation in communication performance. When authentication is required, the perturbation-free representation ( , The input is to the RF fingerprint extraction and matching module, which outputs the authentication result. This invention does not limit the specific RF fingerprint recognition model. The cooperative receiver can input the dedisturbed complex baseband symbol, equivalent channel response, or other fingerprint-bearing features into the RF fingerprint recognition model. , Generally, it includes a feature extraction module and a classification module. The feature extraction module can use convolutional neural networks, recurrent neural networks, or traditional statistical feature extraction methods (such as support vector machines); the classification module can use Softmax classifiers, support vector machines, distance metric matching, or threshold decision authentication, etc.

[0055] This step, through a standardized physical layer transmission process, ensures the compatibility of the RF fingerprint protection mechanism with existing communication protocols. While achieving privacy-protecting signal transmission, it avoids demodulation failures in non-cooperative standard receivers caused by special waveform design, guaranteeing the connectivity and concealment of the communication link. This allows the protection method to be seamlessly integrated into existing wireless communication infrastructure.

[0056] Example 2 like Figure 3 As shown, the present invention proposes a radio frequency fingerprint protection system 10 for a multi-antenna communication system, comprising: The disturbance parameter generation module 100 is used to acquire the communication signal to be transmitted and the pre-shared key, and generate disturbance parameters based on the pre-shared key and the data packet index. The disturbance parameters include a first disturbance factor and a second disturbance factor.

[0057] The weighted processing module 200 is used to perform weighted processing on the training sequence in the communication signal independently according to the transmit antenna dimension using the first perturbation factor, so as to maintain the spatial orthogonal structure between the multi-antenna signals.

[0058] The linkage perturbation processing module 300 is used to perform structured linkage perturbation processing on the data symbols in the communication signal using the first perturbation factor and the second perturbation factor, and to establish a perturbation association between the data symbols and the training sequence.

[0059] The transmitting module 400 is used to transmit the communication signal after the weighting processing and the structured linkage disturbance processing through a wireless channel. The receiving module 500 is used to receive the signal sent by the transmitting module 400 and perform radio frequency fingerprint extraction and matching after decoding.

[0060] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for radio frequency fingerprint protection of a MIMO communication system based on structured perturbations.

[0061] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0062] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for radio frequency fingerprint protection of a MIMO communication system based on structured perturbations, characterized in that, Includes the following steps: S1, obtain the communication signal to be sent and the pre-shared key, and generate perturbation parameters based on the pre-shared key and the data packet index. The perturbation parameters include a first perturbation factor and a second perturbation factor. S2, using the first perturbation factor to perform independent weighting processing on the training sequence in the communication signal according to the transmit antenna dimension, so as to maintain the spatial orthogonal structure between the multi-antenna signals; S3, using the first perturbation factor and the second perturbation factor to perform structured linkage perturbation processing on the data symbols in the communication signal, and establishing a perturbation association between the data symbols and the training sequence; S4, the communication signal after the weighting process and the structured linkage disturbance process is transmitted through the wireless channel.

2. The method as described in claim 1, characterized in that, S1 includes: S11, Generate a data packet index for each data packet to be sent, and the data packet index and the pre-shared key are used together as seed input; S12, using the pre-shared key and the data packet index as input, call a cryptographic pseudo-random function to generate a binary bit stream; S13, the binary bit stream is converted into the first perturbation factor and the second perturbation factor using a mapping function. The first perturbation factor is used for weighted processing of the training sequence, and the second perturbation factor is used for structured linkage perturbation processing of the data symbols.

3. The method as described in claim 2, characterized in that, S13 further includes: S131, the binary bit stream is split into several bit groups, and each bit group is mapped to a perturbed alphabet index; S132, select amplitude parameters and phase parameters from the amplitude set and phase set based on the perturbation alphabet index, wherein the amplitude set is symmetrically designed around the nominal amplitude; S133, construct complex perturbation elements based on the amplitude parameter and the phase parameter, and assemble the complex perturbation elements into the first perturbation factor and the second perturbation factor, wherein the construction formula for a single perturbation element is: ； in, This represents the complex perturbation element. This indicates the amplitude parameter. This represents the phase parameter. It represents the imaginary unit.

4. The method as described in claim 1, characterized in that, S2 includes: S21, Construct a diagonal matrix based on the first perturbation factor, wherein the dimension of the diagonal matrix is ​​adapted to the number of transmit antennas, and wherein the training sequence is a preamble; S22, using the diagonal matrix to... The original preamble matrices on the subcarriers are left-multiplied and weighted to obtain the hidden preamble, where the ... The formula for calculating the hidden preamble on each subcarrier is: ； in, This represents the hidden preamble. Indicates that the first disturbance factor is in the first... Components on each subcarrier This represents the original preamble matrix. This indicates a diagonalization operation.

5. The method as described in claim 1, characterized in that, S3 includes: S31, reuse the first perturbation factor used for the weighting processing of the training sequence as the left multiplication factor of the data symbol; S32, the original data symbols are multiplied element by element using the second perturbation factor to obtain intermediate perturbation data symbols; S33, apply the left multiplication factor to the intermediate perturbation data symbol to obtain the hidden data symbol, where the first... The formula for calculating the hidden data symbols on each subcarrier is: ； in, This represents the hidden data symbol. Indicates that the second disturbance factor is in the first... Components on each subcarrier Represents the original data symbols. This indicates element-wise multiplication.

6. The method as described in claim 1, characterized in that, S4 includes: S41, the hidden preamble and the hidden data symbols are digitally modulated to obtain the time-domain signal; S42, add a cyclic prefix to the time-domain signal to obtain a transmission frame that conforms to the orthogonal frequency division multiplexing standard; S43, the transmitted frame is sent to the wireless channel via the radio frequency front end for reception by cooperative receivers and non-cooperative standard receivers.

7. The method as described in claim 6, characterized in that, S43 further includes: S431, after the cooperative receiver receives the signal sent by the transmitter, it reconstructs the first perturbation factor and the second perturbation factor based on the corresponding key and data packet index; S432, perform corrected channel estimation based on the first perturbation factor, and perform channel equalization and perturbation removal on the received signal to obtain data symbols under noise-limited distortion only; S433 inputs the denoised corrected channel estimation result and the data symbols under noise-limited distortion only into the RF fingerprint extraction and matching module, and outputs the authentication result.

8. The method as described in claim 5, characterized in that, The second perturbation factor used for the structured linkage perturbation processing of the data symbols includes: The second perturbation factor is constructed as a perturbation matrix that matches the dimension of the data symbols; The elements of the perturbation matrix are obtained by bit block mapping and are used to further obfuscate the data segment fingerprint; The element-wise multiplication improves unpredictability to unauthorized parties while maintaining parameter synchronization with cooperative receivers.

9. A radio frequency fingerprint protection system for a MIMO communication system based on structured perturbations, characterized in that, include: The disturbance parameter generation module is used to acquire the communication signal to be transmitted and the pre-shared key, and generate disturbance parameters based on the pre-shared key and the data packet index. The disturbance parameters include a first disturbance factor and a second disturbance factor. The weighting processing module is used to perform independent weighting processing on the training sequence in the communication signal according to the transmit antenna dimension using the first perturbation factor, so as to maintain the spatial orthogonal structure between the multi-antenna signals. The linkage perturbation processing module is used to perform structured linkage perturbation processing on the data symbols in the communication signal using the first perturbation factor and the second perturbation factor, and to establish a perturbation association between the data symbols and the training sequence; The transmitting module is used to transmit the communication signal after the weighting processing and the structured linkage perturbation processing through a wireless channel. The receiving module is used to receive signals sent by the transmitting module and perform radio frequency fingerprint extraction and matching after decoding.

10. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the method as claimed in any one of claims 1-8.

Images