A combination feature determination method and device, electronic equipment and storage medium

Abstract

The application discloses a combination feature determination method and device, electronic equipment and a storage medium. The combination feature determination method comprises the following steps: acquiring radio signals received by each antenna in a synchronous antenna in a current environment; aligning each radio signal to obtain an enhanced enhanced signal; determining a plurality of radio frequency fingerprint features corresponding to the enhanced signal and the weight of each radio frequency fingerprint feature; and combining each radio frequency fingerprint feature to obtain a combination feature in the current environment based on the weight of each radio frequency fingerprint feature. The stability of radio frequency fingerprint feature recognition is solved. Radio signals are received by multiple antennas in a synchronous antenna, and then the radio signals are enhanced. The radio frequency fingerprint features of the radio signals are determined, and then the combination feature is obtained through radio frequency fingerprint feature fusion. The application makes full use of the application scenarios of each radio frequency fingerprint feature, and effectively improves the stability of radio frequency fingerprint feature recognition.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a method, apparatus, electronic device, and storage medium for determining combined features. Background Technology

[0002] Access authentication technology in the Internet of Things (IoT) enables the authentication and authorization of access users, thereby achieving access control for users accessing the local area network (LAN). Common authentication technologies include authentication based on the device's Media Access Control Address (MAC) address, authentication based on security certificates, and authentication based on authentication commands. However, all of these technologies present inherent risks in practical applications. Therefore, a solution based on radio frequency fingerprinting has been proposed.

[0003] However, if the radio frequency fingerprint access authentication scheme is to be truly applied to IoT scenarios, the stability of radio frequency fingerprint features must be addressed. In particular, considering that IoT application scenarios are usually in some complex environments, even higher requirements are placed on stability. Summary of the Invention

[0004] This invention provides a method, apparatus, electronic device, and storage medium for determining combined features to solve the stability problem of radio frequency fingerprint feature recognition.

[0005] According to one aspect of the present invention, a method for determining combined features is provided, comprising:

[0006] Acquire the radio signals received by each antenna in the synchronization antenna under the current environment;

[0007] Align each radio signal to obtain the enhanced signal;

[0008] Determine multiple radio frequency fingerprint features corresponding to the enhanced signal and the weight of each radio frequency fingerprint feature;

[0009] Based on the weight of each radio frequency fingerprint feature, the combined radio frequency fingerprint features are obtained to obtain the combined features under the current environment.

[0010] According to another aspect of the present invention, a combined feature determination apparatus is provided, comprising:

[0011] The acquisition module is used to acquire the radio signals received by each antenna in the synchronization antenna under the current environment;

[0012] The alignment module is used to align each radio signal to obtain the enhanced signal.

[0013] A determination module is used to determine multiple radio frequency fingerprint features corresponding to the enhanced signal and the weight of each radio frequency fingerprint feature;

[0014] The combination module is used to combine the radio frequency fingerprint features based on the weight of each radio frequency fingerprint feature to obtain the combined features under the current environment.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] Synchronization antenna, at least one processor connected to the synchronization antenna; and

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method described in any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the method described in any embodiment of the present invention.

[0020] The technical solution of this invention involves acquiring radio signals received by each antenna in a synchronous antenna under current conditions; aligning each radio signal to obtain an enhanced signal; determining multiple radio frequency fingerprint features corresponding to the enhanced signal and the weights of each radio frequency fingerprint feature; and combining the radio frequency fingerprint features based on their weights to obtain a combined feature under current conditions. This solves the stability problem of radio frequency fingerprint feature recognition. By receiving radio signals through multiple antennas in a synchronous antenna, enhancing the radio signals, determining the radio frequency fingerprint features of the radio signals, and then obtaining a combined feature through radio frequency fingerprint feature fusion, the invention fully utilizes the applicable scenarios of each radio frequency fingerprint feature and effectively improves the stability of radio frequency fingerprint feature recognition.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1This is a flowchart of a method for determining combined features according to an embodiment of the present invention;

[0024] Figure 2 This is a flowchart of another method for determining combined features according to an embodiment of the present invention;

[0025] Figure 3 This is a schematic diagram of a combined feature determination device according to an embodiment of the present invention;

[0026] Figure 4 This is a schematic diagram of the structure of an electronic device that implements the combined feature determination method of the present invention. Detailed Implementation

[0027] 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.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] To address the issue of indistinct RF fingerprint features under low signal-to-noise ratio (SNR), this solution proposes a dual-antenna LoRa (LoRa long-range radio) signal enhancement method. Before acquiring RF fingerprint features, the LoRa signal is first recovered and enhanced. Each RF fingerprint feature has its own applicable scenarios, such as requiring a specific SNR. Therefore, the advantages of various RF fingerprint features can be combined to improve their stability. Under different SNR conditions, by assigning appropriate weights to different RF fingerprint features, a comprehensive RF fingerprint feature is synthesized for identity authentication, thereby improving recognition stability.

[0030] Figure 1This is a flowchart of a combined feature determination method according to an embodiment of the present invention. This embodiment is applicable to the acquisition of radio frequency fingerprint features. The method can be executed by a combined feature determination device, which can be implemented in hardware and / or software. The combined feature determination device can be configured in an electronic device, which can be a receiver. Figure 1 As shown, the method includes:

[0031] S110. Acquire the radio signals received by each antenna in the synchronization antenna under the current environment.

[0032] A synchronization antenna can be considered as multiple antennas on a receiver that synchronously receive radio signals. This embodiment does not limit the number of antennas in the synchronization antenna; for example, it can be a dual-synchronization antenna composed of two antennas. The radio signal can be a LoRa signal, and this is not limited here.

[0033] This step involves synchronously receiving radio signals from each antenna in the synchronization antenna. The electronic device acquires the radio signals received by each antenna in the synchronization antenna. The radio signals can be signals with preambles, such as LoRa signals. Based on the preamble, the corresponding radio frequency fingerprint characteristics of the radio signal can be determined. LoRa signals generally consist of a preamble, synchronization symbols, and data payload, and their specific patterns conform to the general LoRa standard. The preamble consists of 6 to 65535 repeated and identical up-chirp symbols.

[0034] The current environment can be considered the current communication environment. For example, the environment in which the LoRa signal transmitter and electronic devices communicate, including outdoor and indoor environments.

[0035] S120. Align each radio signal to obtain the enhanced signal.

[0036] An enhanced signal can be considered as a signal that has been amplified from a radio signal. This step can align multiple radio signals to obtain an enhanced signal, which then enables subsequent RF fingerprint feature determination based on the enhanced signal.

[0037] This step enhances the signal by aligning the radio signals. Alignment can be based on the phase difference between the radio signals. Then, the multiple radio signals are combined to obtain the enhanced signal.

[0038] S130. Determine the multiple radio frequency fingerprint features corresponding to the enhanced signal and the weight of each radio frequency fingerprint feature.

[0039] Radio frequency fingerprint features are unique characteristics of a device hidden in the radio frequency signals emitted by the transmitter. The main principle behind their generation is that electronic devices, such as terminal devices, will have differences in the production standards and tolerances of various components during the manufacturing process due to factors such as materials or processes. The characteristics that are ultimately reflected in the radio frequency signals by all these factors can serve as the radio frequency fingerprint of the device. These characteristics are unique and cannot be cloned, thus providing a possible solution for identity recognition.

[0040] This step can perform calculations on the enhanced signal to determine multiple radio frequency (RF) fingerprint features. These RF fingerprint features can be one or more of the following: frequency offset features, in-phase quadrature (IQ) offset features, cyclic cross-correlation spectrum features, and cyclic cross-power spectral density features.

[0041] Different radio frequency fingerprint features have different identification methods, and no specific method is specified here.

[0042] After identifying multiple radio frequency fingerprint features, this step determines the weight of each radio frequency fingerprint feature for use in combining radio frequency fingerprint features.

[0043] This step determines the weights of RFID features using a judgment matrix. The elements of the judgment matrix can be composed of the mutual accuracy rates among RFID features. The mutual accuracy rate is the accuracy of a particular RFID feature relative to all RFID features. The relative accuracy rate needs to be determined for all RFID features.

[0044] When determining weights using a judgment matrix, for a target RF fingerprint feature, the weight of that feature can be determined as the ratio of the sum of the target accuracy rates of the target RF fingerprint features to the sum of all elements in the judgment matrix. The target RF fingerprint feature can be considered as the RF fingerprint feature whose weight needs to be determined among multiple RF fingerprint features.

[0045] S140. Based on the weight of each radio frequency fingerprint feature, combine the radio frequency fingerprint features to obtain the combined features under the current environment.

[0046] Combined features can be considered as features obtained by combining radio frequency fingerprint features.

[0047] After determining the weight of each RFID fingerprint feature, this step involves weighted summation of the features to determine the combined feature. The combined feature can be considered as the final extracted RFID fingerprint feature with high stability.

[0048] This embodiment provides a method for determining combined features. It involves acquiring radio signals received by each antenna in a synchronous antenna under current conditions; aligning each radio signal to obtain an enhanced signal; determining multiple radio frequency fingerprint features corresponding to the enhanced signal and the weights of each feature; and combining the features based on their weights to obtain combined features for the current environment. This method solves the stability problem of radio frequency fingerprint feature recognition. By receiving radio signals from multiple antennas in a synchronous antenna, enhancing the signals, determining their radio frequency fingerprint features, and then obtaining combined features through feature fusion, it fully utilizes the applicable scenarios of each feature and effectively improves the stability of radio frequency fingerprint feature recognition.

[0049] Based on the above embodiments, modified embodiments of the above embodiments are proposed. It should be noted that, in order to keep the description brief, only the differences from the above embodiments are described in the modified embodiments.

[0050] In one embodiment, the synchronization antenna is a dual-synchronization antenna, which includes two antennas that receive signals synchronously. Therefore, acquiring the radio signals received by each antenna in the current environment includes:

[0051] Acquire the radio signals received by each antenna in the dual-synchronous antenna system under the current environment.

[0052] Furthermore, the radio signal is a long-range LoRa radio signal.

[0053] In this embodiment, radio signals received by two synchronized antennas within a set time period can be acquired. The radio signals received synchronously by each antenna can be considered as a set of radio signals.

[0054] In one embodiment, aligning each radio signal to obtain the enhanced signal includes:

[0055] Determine the phase difference between the two antennas in the synchronization antenna;

[0056] The two radio signals received synchronously are aligned based on the phase difference;

[0057] The merged and aligned radio signals result in an enhanced signal.

[0058] In this embodiment, when synchronizing radio signals, the phase difference between antennas can be determined first, and then phase rotation can be performed using the phase difference to align the radio signals. After aligning the radio signals, the aligned radio signals can be combined to obtain an enhanced signal. The combining method is not limited here; it could be the summation of the aligned radio signals.

[0059] This embodiment can obtain an enhanced signal based on a set of radio signals, or it can be based on multiple sets of radio signals, determine the enhanced signal corresponding to each set of radio signals, and then determine the average value of each enhanced signal as the final enhanced signal.

[0060] In this embodiment, the phase difference can be determined based on the Fourier transform result of the radio signal.

[0061] In one embodiment, determining the phase difference between the two antennas in the synchronization antenna includes:

[0062] For each group of synchronously received radio signals, the radio signals are multiplied by their conjugate to obtain the multiplied electrical signal;

[0063] Perform a Fourier transform on each of the multiplied electrical signals;

[0064] The phase difference between the two antennas in the synchronization antenna is determined based on the result of the Fourier transform.

[0065] In this embodiment, each set of synchronously received radio signals is multiplied by its conjugate to obtain a multiplied electrical signal. All multiplied electrical signals are then subjected to a Fourier transform. The argument of the frequency corresponding to the peak value of the Fourier transform represents the phase difference between the antennas.

[0066] The following is an exemplary description of a signal enhancement scheme:

[0067] First, the signals (i.e., radio signals) received by the dual synchronous antennas are multiplied by their conjugates to remove channel phase (i.e., for each group of synchronously received radio signals, the radio signals are multiplied by their conjugates to obtain the multiplied electrical signal). Multiplying two radio signals by their conjugates can be understood as dot-multiplying one radio signal (e.g., the radio signal collected by antenna one) by the conjugate of another radio signal (e.g., the radio signal collected by antenna two).

[0068] rx(t) = rx1(t)·rx2(t) * +n(t)

[0069] Here, rx1(t) and rx2(t) represent the LoRa signals received from the same device by the two synchronous antennas, and n(t) is noise, such as Gaussian noise. The channel phase difference between antenna one and antenna two corresponds to the phase of the signal obtained after conjugate multiplication.

[0070] Next, the results rx(t) from several conjugate multiplications are concatenated into a vector and subjected to FFT (i.e., Fourier transform of each multiplied electrical signal). The number of rx(t) should be as small as possible while still being able to obtain RF fingerprint features, thereby improving the utilization rate of the preamble in the received signal. The number of rx(t) varies depending on the signal-to-noise ratio (SNR); for example, at an SNR of -20dB, the number of rx(t) is 8.

[0071] Since the energy of noise does not add up, and the preamble is a repeating LoRa symbol, using multiple rx(t) for FFT aggregates multiple preambles in the FFT window, resulting in clear frequency peaks.

[0072] The argument of the frequency corresponding to the FFT peak is the phase difference between the antennas (i.e., the phase difference between the two antennas in a synchronization antenna is determined based on the result of the Fourier transform):

[0073]

[0074] in, This represents the phase difference between the signals from the first and second antennas, F represents the result of the FFT, max_idx represents the index of the peak value (i.e., the frequency), and angle is the amplitude angle. The result is expressed in radians.

[0075] Averaging the results of multiple symbols yields an enhanced signal:

[0076]

[0077] Where rx1(t) represents the received signal from the first antenna, and rx2(t) represents the received signal from the second antenna. This indicates a phase rotation of rx2(t). Align it with the signal from the first antenna (i.e., align the two radio signals received synchronously based on the phase difference). After combining the weak signals from multiple antennas (i.e., combining the aligned radio signals to obtain the enhanced signal), the enhanced signal (i.e., Y(t)) is obtained.

[0078] In one embodiment, determining the plurality of radio frequency fingerprint features corresponding to the enhanced signal includes:

[0079] After normalizing the enhanced signal, a normalized signal is obtained.

[0080] Determine the instantaneous phase corresponding to the preamble in the normalized signal;

[0081] The frequency of each sampling point in the preamble is determined based on the instantaneous phase;

[0082] The average value of each frequency is determined as the coarse frequency offset;

[0083] In the preamble, a differential operation is performed every symbol to obtain the average frequency offset value as the fine frequency offset;

[0084] The sum of the coarse frequency offset and the fine frequency offset is determined as the frequency offset feature, which is used as a radio frequency fingerprint feature.

[0085] A normalized signal can be considered as the signal after normalization processing. Frequency offset features can be considered as the extracted features that reflect frequency offset.

[0086] Normalizing the enhanced signal can be considered as multiplying the enhanced signal by a normalization factor.

[0087] This embodiment can determine the instantaneous phase corresponding to the preamble in the normalized signal based on the I-channel and Q-channel signals in the normalized signal. The instantaneous phase corresponding to the preamble can be considered as the instantaneous phase of all sampled signals of the preamble.

[0088] When determining the frequency of each sampling point, it can be based on the phase and sampling frequency of adjacent sampling points.

[0089] The following is an example of how to determine frequency offset characteristics:

[0090] First, establish the signal transmission model. The LoRa transmitter's I and Q signals, after modulation, can be represented as follows:

[0091] s(t)=I(t)cosω snd t+Q(t)sinω snd t

[0092] Where, ω snd Let I(t) represent the instantaneous frequency of the signal, and Q(t) represent the components of the transmitted signal s(t) in the I and Q paths, respectively. The enhanced signal also removes channel interference, introducing only the carrier frequency offset, and can be expressed as:

[0093] Y(t)=s(t)e jΔωt

[0094] Here, Δω represents the carrier frequency offset in the enhanced signal, which can be obtained by multiplying the transmitted signal s(t) by the rotation factor e in the complex plane. jΔωt The final enhanced signal obtained by the receiver in the frequency domain is:

[0095] r(n) = I(n) + jQ(n)

[0096] I(n) represents the component of the signal in the I path, Q(n) represents the component of the signal in the Q path, n represents the nth sampling point, and n takes values ​​from 1 to N, where N is the number of sampling points for a preamble symbol in the received signal.

[0097] Optionally, the frequency offset characteristic calculation stage includes:

[0098] Step 1: To better extract frequency offset features, the enhanced signal r(n) obtained from the receiver in the frequency domain needs to be preprocessed. Specifically, normalization is performed by multiplying r(n) by the normalization factor N. r ,in

[0099]

[0100] Step 2: Coarse frequency offset extraction to determine the instantaneous phase corresponding to the preamble in the normalized signal:

[0101]

[0102] Calculate the frequency of each sampling point (i.e., determine the frequency of each sampling point in the preamble based on the instantaneous phase):

[0103]

[0104] Among them, f s This represents the sampling frequency. Then, calculating the average frequency of each sampling point in the preamble yields the coarse frequency offset Δf. 粗 (That is, the average value of each frequency is determined as the coarse frequency offset).

[0105] Step 3: Calculate the fine frequency offset. Specifically, perform a differential operation on the preamble every other symbol and take the average frequency offset value (i.e., perform a differential operation on the preamble every other symbol and take the average frequency offset value as the fine frequency offset). The formula is:

[0106]

[0107] Finally, the coarse frequency offset plus the fine frequency offset equals the carrier frequency offset feature. That is, the sum of the coarse and fine frequency offsets is determined as the frequency offset feature, which serves as a type of radio frequency fingerprint feature.

[0108] Δf=Δf 精 +Δf 粗

[0109] Optionally, determining the multiple radio frequency fingerprint features corresponding to the enhanced signal includes:

[0110] The enhanced signal is then frequency offset compensated.

[0111] Plot the in-phase and quadrature signals in the compensated signal on a two-dimensional plane;

[0112] The signal on the two-dimensional plane is divided into multiple equal parts;

[0113] The average value of each part of the signal is obtained by averaging the signals of each part.

[0114] Based on the average value of each segment and the theoretical center point on the unit circle, the in-phase orthogonal offset feature of each segment is determined, and the in-phase orthogonal offset feature is used as a radio frequency fingerprint feature.

[0115] In-phase signals can be considered as I-channel signals. Quadrature signals can be considered as Q-channel signals. In this embodiment, the two-dimensional plane can be divided into eight equal parts.

[0116] The steps for determining the IQ shift feature include:

[0117] Step 1: Perform frequency offset compensation on the enhanced signal, and then plot the obtained I-path signal and Q-path signal on a two-dimensional plane (that is, plot the in-phase signal and quadrature signal in the compensated signal on a two-dimensional plane), and divide it into 8 equal parts at 45° intervals (that is, divide the signal on the two-dimensional plane into multiple equal parts).

[0118] Step 2: Sum the sampling points in each equal sector and take the mean to obtain C. z This involves averaging the signal across each equal segment to obtain the average value of each segment. The IQ offset characteristic is the average value C of each sector. z The difference between the theoretical center point C on the unit circle (the unit circle is defined by the points where the I-channel and Q-channel signals are plotted on a two-dimensional plane, with a radius of one unit) and the unit circle. Specifically, based on the average value of each equal part and the theoretical center point on the unit circle, the in-phase orthogonal offset characteristics of each equal part are determined as follows:

[0119] z represents the z-th sector.

[0120] Optionally, determining the multiple radio frequency fingerprint features corresponding to the enhanced signal includes:

[0121] Arrange the preamble symbols in each of the enhanced signals in chronological order;

[0122] Starting from the u-th preamble symbol in each of the preamble symbols, determine the cepstrum of each preamble symbol;

[0123] The cepstral spectra of two temporally adjacent preamble symbols are subjected to cyclic cross-correlation to obtain a cyclic cross-correlation spectrum, which serves as a radio frequency fingerprint feature.

[0124] u can be a positive integer, such as 8.

[0125] Find the characteristic spectrum of cyclic cross-correlation:

[0126] Step 1: For the enhanced preamble symbols arranged in time sequence, to prevent frequency jumps at the beginning and end of the FM chirp signal from affecting the frequency distribution characteristics, starting from the 8th symbol, the cepstral is calculated first. That is, starting from the u-th preamble symbol, the cepstral of each preamble symbol is determined:

[0127] C i (m)=F -1 {log(F(X i (t)))}

[0128] Among them, X i (t) represents the preamble symbol in the time domain representation.

[0129] Step 2: Cepstrum of two adjacent preamble symbols (the cepstrum improves the quality of the preamble symbols): C i (m),C i+1 (m) Perform cyclic cross-correlation operations:

[0130]

[0131] Among them, ((nm)) N This indicates a cyclic shift of nm sampling points.

[0132] Step 3: Repeat the calculation multiple times to obtain multiple cyclic cross-correlation spectra.

[0133] Optionally, determining the multiple radio frequency fingerprint features corresponding to the enhanced signal includes:

[0134] Arrange the preamble symbols in each of the enhanced signals in chronological order;

[0135] Starting from the u-th preamble symbol in each of the preamble symbols, determine the cyclic cross-correlation function for each preamble symbol;

[0136] Based on the cyclic cross-correlation function, the cross-power spectral density vector of each preamble symbol is determined;

[0137] The mean of each v cross-power spectral density vector is determined as the cyclic cross-power spectral density feature, which serves as a radio frequency fingerprint feature.

[0138] u and v can be positive integers, such as 8.

[0139] Determining the characteristics of the cyclic cross power spectral density includes:

[0140] Step 1: Calculate the cross-power spectral density vector G for each preamble symbol. i That is, based on the cyclic cross-correlation function, the cross-power spectral density vector of each preamble symbol is determined:

[0141]

[0142] Where k is the independent variable of the cross power spectral density vector.

[0143] Step 2: Calculate the mean of every 8 cross power spectral density vectors to obtain the cyclic cross power spectral density feature.

[0144] Optionally, determining the weight of each of the radio frequency fingerprint features includes:

[0145] The weight of the radio frequency fingerprint feature is determined by the ratio of the sum of the elements corresponding to the same radio frequency fingerprint feature in the judgment matrix to the order of the judgment matrix.

[0146] The elements of the judgment matrix include the relative accuracy of each radio frequency fingerprint feature in the current environment. The relative accuracy of the radio frequency fingerprint feature is the accuracy of the radio frequency fingerprint feature relative to all radio frequency fingerprint features. The elements corresponding to the same radio frequency fingerprint feature include the relative accuracy of the same radio frequency fingerprint feature relative to all radio frequency fingerprint features.

[0147] The judgment matrix can be considered as a matrix used to determine the weights of RFID fingerprint features. Relative accuracy can be considered as the accuracy of the current RFID fingerprint feature relative to all other RFID fingerprint features. The accuracy of each RFID fingerprint feature is obtained using the SVW, CNN, K-Nearest Neighbors, and Naive Bayes algorithms of the classification learner, and the highest accuracy value among the four algorithms is denoted as px or py. In this embodiment, the elements corresponding to the same RFID fingerprint feature are summed and divided by the order in the judgment matrix; this summation is used as the weight of that RFID fingerprint feature.

[0148] When determining weights, a hierarchical model can be used, which consists of three levels: the target layer, the criterion layer, and the execution layer. The target layer represents the purpose of the decision, i.e., the weighted comprehensive RF fingerprint features; the criterion layer represents the factors considered and the decision criteria, indicating the performance of the four features under different signal-to-noise ratios; the execution layer represents the alternative solutions during the decision-making process, i.e., frequency offset features, IQ offset features, cyclic cross-correlation spectrum, and cyclic cross-power spectral density features.

[0149] The relative accuracy in this embodiment can be scaled from 1 to 9. Table 1 is a schematic table of a scaling method provided by an embodiment of the present invention.

[0150] Table 1. A scale illustration provided by an embodiment of the present invention.

[0151] 1 The accuracy rates are the same for both factors. 3 Between the two factors, the former is slightly more accurate than the latter. 5 Between the two factors, the former is more accurate than the latter. 7 Compared to the latter, the former is far more accurate. 9 Between the two factors, the former is absolutely more accurate than the latter. 2,4,6,8 The intermediate value between the above scales reciprocal If A's scale relative to B is 3, then B's scale relative to A is 1 / 3.

[0152] The scale value is determined by the relative accuracy, p x ,p yLet each represent the accuracy rate of a fingerprint feature. Then, the relative accuracy rate can be expressed as:

[0153]

[0154] p xy It can be considered as p x The corresponding fingerprint features relative to p y The relative accuracy of fingerprint features.

[0155] Generally, an accuracy rate below 80% is considered an invalid feature. In this case, any valid feature is scaled to 9 relative to it. The relative accuracy of valid features is divided into nine scales, and the comparison table of the nine scales is shown in Table 2. Table 2 is a comparison table of scales and relative accuracy provided by an embodiment of the present invention.

[0156] Table 2. Comparison table of scale and relative accuracy provided in the embodiments of the present invention.

[0157] 0％ 1 0%≤pxy<3.125% 2 3.125%≤pxy<6.25% 3 6.25%≤pxy<9.375% 4 9.375%≤pxy<12.5% 5 12.5%≤pxy<15.625% 6 15.625%≤pxy<18.75% 7 18.75%≤pxy<21.875% 8 21.875% ≤ pxy ≤ 25% 9

[0158] Judgment matrix A was obtained through experiments:

[0159]

[0160] Where o, g, l, d represent frequency offset characteristics, IQ offset characteristics, cyclic cross-correlation spectrum, and cyclic cross-power spectral density characteristics, respectively, and a og The scale representing the frequency offset feature relative to the IQ offset feature, and a og =1 / a go And so on.

[0161] The eigenvectors, eigenvalues, and weights are calculated based on the judgment matrix.

[0162] First, normalize the judgment matrix column by column:

[0163]

[0164] Where l1 represents the sum of the first column of the judgment matrix A, such as l1 = 1 + a go +a lo +a do l2 represents the sum of the second column of matrix A, l3 represents the sum of the third column of matrix A, and l4 represents the sum of the fourth column of matrix A.

[0165] The resulting matrix is ​​then summed row-wise (i.e., the sum of elements in the judgment matrix corresponding to the same RF fingerprint feature). Each element of the resulting vector is then divided by the degree of the judgment matrix to obtain the weight vector. The weight w of a specific RF fingerprint feature is then calculated. x Represented as:

[0166]

[0167] Then the consistency test (CI) is calculated.

[0168]

[0169] Where degree represents the order of the matrix, λ max This represents the largest eigenvalue of the matrix.

[0170] The consistency ratio (CR) is calculated, where the consistency index (RI) is directly obtained by looking up the random consistency RI table. The random consistency RI table includes RI values ​​corresponding to different matrix orders, with each matrix order uniquely corresponding to a single RI value.

[0171] CR = CI / RI

[0172] If the CR value is less than 0.1, it means the consistency test has passed; otherwise, it means the consistency test has not passed. If the data fails the consistency test, it is necessary to check for logical problems and re-analyze.

[0173] Finally, the comprehensive radio frequency fingerprint features were obtained.

[0174] After passing the consistency check, the final combined RF fingerprint features in the current environment are obtained by combining the obtained features and weight allocation relationships (i.e., based on the weight of each RF fingerprint feature, the combined RF fingerprint features in the current environment are obtained by combining the RF fingerprint features).

[0175] The combined feature determination method provided by this invention can be considered a method for acquiring comprehensive features of dual-antenna low signal-to-noise ratio LoRa radio frequency fingerprints. To overcome the shortcomings of existing solutions, a single radio frequency fingerprint feature cannot meet the needs of security identification. Different fingerprint features are typically suitable for different scenarios or channel conditions. To combine the advantages of these features, they can be combined. Specifically, different features can be assigned different weights in different scenarios. For example, in low signal-to-noise ratio scenarios, features suitable for low signal-to-noise ratio conditions are prioritized, and similar approaches are taken in other scenarios. These features are then combined to form a comprehensive feature. This comprehensive feature can fully utilize the applicable scenarios of each feature, maximize the stability of the fingerprint, effectively increase the feature dimensionality, and better adapt to the needs of real-world scenarios.

[0176] To achieve the above objectives, this invention proposes a dual-antenna method for acquiring comprehensive features of LoRa radio frequency fingerprints under low signal-to-noise ratio conditions, used for extracting highly stable radio frequency fingerprints in IoT scenarios, comprising the following steps:

[0177] Step 1: Signal enhancement (i.e., acquiring the radio signals received by each antenna in the synchronization antenna under the current environment; aligning each radio signal to obtain the enhanced signal).

[0178] Step 2: Calculate the frequency offset feature (i.e., normalize the enhanced signal to obtain a normalized signal; determine the instantaneous phase corresponding to the preamble in the normalized signal; determine the frequency of each sampling point in the preamble based on the instantaneous phase; determine the average value of each frequency as the coarse frequency offset; perform a differential operation in the preamble every symbol interval to obtain the average frequency offset value as the fine frequency offset; determine the sum of the coarse frequency offset and the fine frequency offset as the frequency offset feature, which is used as a radio frequency fingerprint feature).

[0179] Step 3: Calculate the IQ offset features (i.e., perform frequency offset compensation on the enhanced signal; plot the in-phase and quadrature signals in the compensated signal on a two-dimensional plane; divide the signal on the two-dimensional plane into multiple equal parts; average the signal of each part to obtain the average value of each part; based on the average value of each part and the theoretical center point on the unit circle, determine the in-phase and quadrature offset features of each part, which serve as a radio frequency fingerprint feature).

[0180] Step 4: Calculate the cyclic cross-correlation spectrum (i.e., arrange the preamble symbols in each of the enhanced signals in time sequence; starting from the u-th preamble symbol in each of the preamble symbols, determine the cepstral of each preamble symbol; perform cyclic cross-correlation operation on the cepstral of two temporally adjacent preamble symbols to obtain the cyclic cross-correlation spectrum, which serves as a radio frequency fingerprint feature).

[0181] Step 5: Calculate the cyclic cross-power spectral density feature (i.e., arrange the preamble symbols in each of the enhanced signals in time sequence; starting from the u-th preamble symbol in each of the preamble symbols, determine the cyclic cross-correlation function of each preamble symbol; based on the cyclic cross-correlation function, determine the cross-power spectral density vector of each preamble symbol; determine the mean of every m cross-power spectral density vectors as the cyclic cross-power spectral density feature, which serves as a radio frequency fingerprint feature).

[0182] Step 6: Calculate and obtain comprehensive features based on environmental conditions (i.e., combine the RF fingerprint features based on the weight of each RF fingerprint feature to obtain the combined features under the current environment).

[0183] Figure 2 This is a flowchart of another method for determining combined features according to an embodiment of the present invention. Figure 2 The process for determining the features is shown: signal enhancement, determination of frequency offset features, determination of IQ offset features, determination of cyclic cross-correlation features, determination of cross-power spectrum features, and acquisition of comprehensive features.

[0184] Signal enhancement includes conjugate multiplication of dual-antenna signals (i.e., for each group of synchronously received radio signals, the radio signals are conjugate multiplied to obtain the multiplied electrical signal), multiple results are aggregated for FFT calculation (i.e., Fourier transform is performed on each of the multiplied electrical signals), phase difference is calculated (i.e., the phase difference between the two antennas in the synchronization antenna is determined based on the Fourier transform result), and phase difference is compensated (i.e., the two synchronously received radio signals are aligned based on the phase difference), and the enhanced signal is obtained by averaging (i.e., the aligned radio signals are combined to obtain the enhanced signal).

[0185] The frequency offset features include signal normalization preprocessing (i.e., normalizing the enhanced signal to obtain a normalized signal), coarse frequency offset extraction (i.e., determining the instantaneous phase corresponding to the preamble in the normalized signal; determining the frequency of each sampling point in the preamble based on the instantaneous phase; and determining the average value of each frequency as the coarse frequency offset), and fine frequency offset extraction (i.e., performing a differential operation every symbol interval in the preamble and taking the average frequency offset value as the fine frequency offset).

[0186] The determination of IQ offset features includes signal frequency offset compensation (i.e., frequency offset compensation of the enhanced signal) and calculation of IQ offset features (i.e., plotting the in-phase and quadrature signals in the compensated signal on a two-dimensional plane; dividing the signal on the two-dimensional plane into multiple equal parts; averaging the signal of each equal part to obtain the average value of each equal part; determining the in-phase and quadrature offset features of each equal part based on the average value of each equal part and the theoretical center point on the unit circle, wherein the in-phase and quadrature offset features are used as a radio frequency fingerprint feature).

[0187] The determination of cyclic cross-correlation features includes calculating the cepstrum (i.e., arranging the preamble symbols in each enhanced signal in time sequence; starting from the u-th preamble symbol in each preamble symbol, determining the cepstrum of each preamble symbol) and calculating cyclic cross-correlation (i.e., performing cyclic cross-correlation operation on the cepstrums of two temporally adjacent preamble symbols to obtain the cyclic cross-correlation spectrum, which serves as a radio frequency fingerprint feature).

[0188] The determination of cross-power spectral density features includes calculating the cross-power spectral density vector (i.e., arranging the preamble symbols in each of the enhanced signals in time sequence; determining the cyclic cross-correlation function of each preamble symbol starting from the u-th preamble symbol; and determining the cross-power spectral density vector of each preamble symbol based on the cyclic cross-correlation function) and calculating the mean (i.e., determining the mean of every m cross-power spectral density vectors as the cyclic cross-power spectral density feature, which serves as a radio frequency fingerprint feature).

[0189] Obtaining comprehensive features includes establishing a hierarchical structure model, determining the scale, constructing a judgment matrix, and performing consistency checks.

[0190] This invention proposes a method for acquiring comprehensive LoRa radio frequency fingerprint features, including signal enhancement, frequency offset features, IQ offset features, cepstral cyclic cross-correlation features, cyclic cross-power spectral density features, and a comprehensive feature calculation stage based on environmental conditions. Weights are assigned to the four radio frequency fingerprint features according to specific environmental conditions, making the comprehensive radio frequency fingerprint features more reliable. This method fully utilizes the advantages of multiple existing radio frequency fingerprint features, can adapt to the needs of different signal-to-noise ratios in IoT environments, effectively improves the comprehensive stability of radio frequency fingerprint features, and enhances the implementation performance of radio frequency fingerprint authentication schemes.

[0191] Figure 3 This is a schematic diagram of a combined feature determination device according to an embodiment of the present invention, as shown below. Figure 3 As shown, the device includes:

[0192] The acquisition module 310 is used to acquire the radio signals received by each antenna in the synchronization antenna under the current environment;

[0193] Alignment module 320 is used to align each radio signal to obtain an enhanced signal;

[0194] The determination module 330 is used to determine multiple radio frequency fingerprint features corresponding to the enhanced signal and the weight of each radio frequency fingerprint feature;

[0195] The combination module 340 is used to combine the radio frequency fingerprint features based on the weight of each radio frequency fingerprint feature to obtain the combined features in the current environment.

[0196] Optionally, the synchronization antenna is a dual synchronization antenna.

[0197] In one embodiment, the radio signal is a long-range LoRa radio signal.

[0198] Optional, alignment module 320 includes:

[0199] The determining unit is used to determine the phase difference between the two antennas in the synchronization antenna;

[0200] An alignment unit is used to align two synchronously received radio signals based on the phase difference.

[0201] The merging unit is used to merge aligned radio signals to obtain an enhanced signal.

[0202] Optional, define the unit, specifically for:

[0203] For each group of synchronously received radio signals, the radio signals are multiplied by their conjugate to obtain the multiplied electrical signal;

[0204] Perform a Fourier transform on each of the multiplied electrical signals;

[0205] The phase difference between the two antennas in the synchronization antenna is determined based on the result of the Fourier transform.

[0206] Optionally, module 330 is determined, specifically for:

[0207] After normalizing the enhanced signal, a normalized signal is obtained.

[0208] Determine the instantaneous phase corresponding to the preamble in the normalized signal;

[0209] The frequency of each sampling point in the preamble is determined based on the instantaneous phase;

[0210] The average value of each frequency is determined as the coarse frequency offset;

[0211] In the preamble, a differential operation is performed every symbol to obtain the average frequency offset value as the fine frequency offset;

[0212] The sum of the coarse frequency offset and the fine frequency offset is determined as the frequency offset feature, which is used as a radio frequency fingerprint feature.

[0213] Optionally, module 330 is determined, specifically for:

[0214] The enhanced signal is then frequency offset compensated.

[0215] Plot the in-phase and quadrature signals in the compensated signal on a two-dimensional plane;

[0216] The signal on the two-dimensional plane is divided into multiple equal parts;

[0217] The average value of each part of the signal is obtained by averaging the signals of each part.

[0218] Based on the average value of each segment and the theoretical center point on the unit circle, the in-phase orthogonal offset feature of each segment is determined, and the in-phase orthogonal offset feature is used as a radio frequency fingerprint feature.

[0219] Optionally, module 330 is determined, specifically for:

[0220] Arrange the preamble symbols in each of the enhanced signals in chronological order;

[0221] Starting from the u-th preamble symbol in each of the preamble symbols, determine the cepstrum of each preamble symbol;

[0222] The cepstral spectra of two temporally adjacent preamble symbols are subjected to cyclic cross-correlation to obtain a cyclic cross-correlation spectrum, which serves as a radio frequency fingerprint feature.

[0223] Optionally, module 330 is determined, specifically for:

[0224] Arrange the preamble symbols in each of the enhanced signals in chronological order;

[0225] Starting from the u-th preamble symbol in each of the preamble symbols, determine the cyclic cross-correlation function for each preamble symbol;

[0226] Based on the cyclic cross-correlation function, the cross-power spectral density vector of each preamble symbol is determined;

[0227] The mean of each v cross-power spectral density vector is determined as the cyclic cross-power spectral density feature, which serves as a radio frequency fingerprint feature.

[0228] Optionally, module 330 is determined, specifically for:

[0229] The weight of the radio frequency fingerprint feature is determined by the ratio of the sum of the elements corresponding to the same radio frequency fingerprint feature in the judgment matrix to the order of the judgment matrix.

[0230] The elements of the judgment matrix include the relative accuracy of each radio frequency fingerprint feature in the current environment. The relative accuracy of the radio frequency fingerprint feature is the accuracy of the radio frequency fingerprint feature relative to all radio frequency fingerprint features. The elements corresponding to the same radio frequency fingerprint feature include the relative accuracy of the same radio frequency fingerprint feature relative to all radio frequency fingerprint features.

[0231] The combined feature determination device provided in the embodiments of the present invention can execute the combined feature determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0232] Figure 4 This is a schematic diagram of the structure of an electronic device implementing the combined feature determination method of embodiments of the present invention. Electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. Electronic device 10 may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0233] like Figure 4 As shown, the electronic device 10 includes a synchronization antenna (not shown), at least one processor 11 connected to the synchronization antenna, and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores a computer program executable by the at least one processor 11, which enables the at least one processor 11 to perform the method provided by the present invention. The synchronization antenna is used to receive radio signals.

[0234] The processor 11 can perform various appropriate actions and processes based on a computer program stored in the read-only memory (ROM) 12 or a computer program loaded from the storage unit 18 into the random access memory (RAM) 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0235] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0236] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processors (DSPs), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as combined feature determination methods.

[0237] In some embodiments, the combined feature determination method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or mounted on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the combined feature determination method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the combined feature determination method by any other suitable means (e.g., by means of firmware).

[0238] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard parts (ASSPs), systems-on-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0239] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0240] In the context of this invention, a computer-readable storage medium stores computer instructions that are used to cause a processor to execute and implement the combined feature determination method provided by this invention.

[0241] Computer-readable storage media can be tangible media that may contain or store computer programs for use by or in conjunction with an instruction execution system, apparatus, or device. Computer-readable storage media can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0242] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a cathode ray tube (CRT) or a liquid crystal display (LCD)) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0243] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0244] A computing system can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is established by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system. It addresses the shortcomings of traditional physical hosts and Virtual Private Server (VPS) services, such as high management difficulty and weak business scalability.

[0245] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0246] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for determining combined features, characterized in that, include: Acquire the radio signals received by each antenna in the synchronization antenna under the current environment; Align each radio signal to obtain the enhanced signal; Determine multiple radio frequency fingerprint features corresponding to the enhanced signal and the weight of each radio frequency fingerprint feature; Based on the weight of each radio frequency fingerprint feature, the combined radio frequency fingerprint features are obtained to obtain the combined features in the current environment; The determination of the weight of each of the radio frequency fingerprint features includes: The weight of the radio frequency fingerprint feature is determined by the ratio of the sum of the elements corresponding to the same radio frequency fingerprint feature in the judgment matrix to the order of the judgment matrix. The elements of the judgment matrix include the relative accuracy of each of the radio frequency fingerprint features in the current environment, where the relative accuracy of the radio frequency fingerprint features is the accuracy of each radio frequency fingerprint feature relative to all radio frequency fingerprint features.

2. The method according to claim 1, characterized in that, The synchronization antenna is a dual synchronization antenna.

3. The method according to claim 1, characterized in that, The radio signal is a long-range LoRa radio signal.

4. The method according to claim 1, characterized in that, The alignment of each radio signal to obtain the enhanced signal includes: Determine the phase difference between the two antennas in the synchronization antenna; The two radio signals received synchronously are aligned based on the phase difference; The merged and aligned radio signals result in an enhanced signal.

5. The method according to claim 4, characterized in that, Determining the phase difference between the two antennas in the synchronization antenna includes: For each group of synchronously received radio signals, the radio signals are multiplied by their conjugate to obtain the multiplied electrical signal; Perform a Fourier transform on each of the multiplied electrical signals; The phase difference between the two antennas in the synchronization antenna is determined based on the result of the Fourier transform.

6. The method according to claim 1, characterized in that, The determination of the multiple radio frequency fingerprint features corresponding to the enhanced signal includes: After normalizing the enhanced signal, a normalized signal is obtained. Determine the instantaneous phase corresponding to the preamble in the normalized signal; The frequency of each sampling point in the preamble is determined based on the instantaneous phase; The average value of each frequency is determined as the coarse frequency offset; In the preamble, a differential operation is performed every symbol to obtain the average frequency offset value as the fine frequency offset; The sum of the coarse frequency offset and the fine frequency offset is determined as the frequency offset feature, which is used as a radio frequency fingerprint feature.

7. The method according to claim 1, characterized in that, The determination of the multiple radio frequency fingerprint features corresponding to the enhanced signal includes: The enhanced signal is then frequency offset compensated. Plot the in-phase and quadrature signals in the compensated signal on a two-dimensional plane; The signal on the two-dimensional plane is divided into multiple equal parts; The average value of each part of the signal is obtained by averaging the signals of each part. Based on the average value of each segment and the theoretical center point on the unit circle, the in-phase orthogonal offset feature of each segment is determined, and the in-phase orthogonal offset feature is used as a radio frequency fingerprint feature.

8. The method according to claim 1, characterized in that, The determination of the multiple radio frequency fingerprint features corresponding to the enhanced signal includes: Arrange the preamble symbols in each of the enhanced signals in chronological order; Starting from the u-th preamble symbol in each of the preamble symbols, determine the cepstrum of each preamble symbol; The cepstral spectra of two temporally adjacent preamble symbols are subjected to cyclic cross-correlation to obtain a cyclic cross-correlation spectrum, which serves as a radio frequency fingerprint feature.

9. The method according to claim 1, characterized in that, The determination of the multiple radio frequency fingerprint features corresponding to the enhanced signal includes: Arrange the preamble symbols in each of the enhanced signals in chronological order; Starting from the u-th preamble symbol in each of the preamble symbols, determine the cyclic cross-correlation function for each preamble symbol; Based on the cyclic cross-correlation function, the cross-power spectral density vector of each preamble symbol is determined; The mean of each v cross-power spectral density vector is determined as the cyclic cross-power spectral density feature, which serves as a radio frequency fingerprint feature.

10. A device for determining combined features, characterized in that, include: The acquisition module is used to acquire the radio signals received by each antenna in the synchronization antenna under the current environment; The alignment module is used to align each radio signal to obtain the enhanced signal. A determination module is used to determine multiple radio frequency fingerprint features corresponding to the enhanced signal and the weight of each radio frequency fingerprint feature; The combination module is used to combine the radio frequency fingerprint features based on the weight of each radio frequency fingerprint feature to obtain the combined features in the current environment; The determining module is specifically used for: The weight of the radio frequency fingerprint feature is determined by the ratio of the sum of the elements corresponding to the same radio frequency fingerprint feature in the judgment matrix to the order of the judgment matrix. The elements of the judgment matrix include the relative accuracy of each of the radio frequency fingerprint features in the current environment, where the relative accuracy of the radio frequency fingerprint features is the accuracy of each radio frequency fingerprint feature relative to all radio frequency fingerprint features.

11. An electronic device, characterized in that, The electronic device includes: Synchronization antenna, at least one processor connected to the synchronization antenna; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that are used to cause a processor to execute the method of any one of claims 1-9.

Images