Communication radiation source identification method and device, electronic equipment and medium

Abstract

The application provides a communication radiation source identification method and device, electronic equipment and medium. The method comprises the following steps: determining a statistical feature of a Hilbert marginal spectrum corresponding to a to-be-identified radiation source signal based on the Hilbert marginal spectrum, the statistical feature being used to indicate a fingerprint feature of a communication radiation source, and the statistical feature comprising a mean value and a variance of the Hilbert marginal spectrum; inputting the statistical feature into a support vector machine classifier to obtain a radiation source category corresponding to the to-be-identified radiation source signal. The communication radiation source identification method provided by the application can greatly reduce the overall calculation complexity of the algorithm.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method, apparatus, electronic device, and medium for identifying communication radiation sources. Background Technology

[0002] Individual identification of communication radiation sources is a technique that identifies a specific individual that emits a given radio signal by receiving and measuring the radio signals emitted by an individual communication radiation source and extracting the individual's unique fingerprint information from the radio signals.

[0003] In recent years, deep learning methods have been increasingly widely used in the field of communication radiation source identification. Deep learning-based communication radiation source identification mainly involves designing neural networks for feature extraction and classifying individual communication radiation sources. While using deep learning methods in classification can improve identification accuracy, it also increases the overall computational complexity of the algorithm to some extent. Summary of the Invention

[0004] This invention provides a method, apparatus, electronic device, and medium for identifying communication radiation sources, in order to solve the problem of high complexity in existing communication radiation source identification algorithms.

[0005] This invention provides a method for identifying communication radiation sources, comprising:

[0006] Based on the Hilbert marginal spectrum corresponding to the signal of the radiation source to be identified, the statistical characteristics of the Hilbert marginal spectrum are determined. The statistical characteristics are used to indicate the fingerprint characteristics of the communication radiation source. The statistical characteristics include the mean and variance of the Hilbert marginal spectrum.

[0007] The statistical features are input into a support vector machine classifier to obtain the radiation source category corresponding to the radiation source signal to be identified.

[0008] In some embodiments, the support vector machine classifier is obtained in the following manner:

[0009] Obtain the Hilbert marginal spectrum samples corresponding to the radiation source signal samples;

[0010] Determine the mean and variance of the Hilbert marginal spectrum sample;

[0011] The support vector machine classifier is trained based on the radiation source category corresponding to the radiation source signal sample, and the mean and variance of the Hilbert marginal spectrum sample.

[0012] In some embodiments, before determining the statistical characteristics of the Hilbert marginal spectrum based on the Hilbert marginal spectrum corresponding to the signal of the radiation source to be identified, the method further includes:

[0013] Acquire the signal of the radiation source to be identified;

[0014] Perform a Hilbert-Huang transform on the radiation source signal to be identified to obtain the Hilbert time spectrum corresponding to the radiation source signal to be identified;

[0015] Based on the Hilbert time spectrum corresponding to the radiation source signal to be identified, the Hilbert marginal spectrum corresponding to the radiation source signal to be identified is determined.

[0016] In some embodiments, after training the support vector machine classifier, the method further includes:

[0017] Based on the hardware damage parameters of the communication radiation source corresponding to the statistical characteristics, a sensitivity analysis is performed on the trained support vector machine classifier to determine the sensitive parameters that affect the classification and recognition performance.

[0018] The hardware impairment parameters include: filter distortion parameters, IQ imbalance gain imbalance parameters, IQ imbalance phase imbalance parameters, and different order coefficients of the power amplifier; the sensitive parameters include: IQ imbalance gain imbalance parameters, first-order coefficient of power amplifier nonlinear error, and third-order coefficient of power amplifier nonlinear error.

[0019] In some embodiments, training the support vector machine classifier includes:

[0020] The support vector machine classifier is optimized based on the 5-fold cross-validation method.

[0021] The present invention also provides a communication radiation source identification device, comprising:

[0022] The determination module is used to determine the statistical characteristics of the Hilbert marginal spectrum based on the Hilbert marginal spectrum corresponding to the signal of the radiation source to be identified. The statistical characteristics are used to indicate the fingerprint characteristics of the communication radiation source. The statistical characteristics include the mean and variance of the Hilbert marginal spectrum.

[0023] The classification module is used to input the statistical features into a support vector machine classifier to obtain the radiation source category corresponding to the radiation source signal to be identified.

[0024] In some embodiments, the apparatus further includes:

[0025] The training module is used to obtain Hilbert marginal spectrum samples corresponding to radiation source signal samples;

[0026] Determine the mean and variance of the Hilbert marginal spectrum sample;

[0027] The support vector machine classifier is trained based on the radiation source category corresponding to the radiation source signal sample, and the mean and variance of the Hilbert marginal spectrum sample.

[0028] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the communication radiation source identification method as described above.

[0029] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the communication radiation source identification method as described above.

[0030] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the communication radiation source identification method as described above.

[0031] The communication radiation source identification method, device, electronic device, and medium provided by this invention determine the statistical characteristics of the Hilbert marginal spectrum corresponding to the radiation source signal to be identified, and use the mean and variance in the statistical characteristics as fingerprint features of the communication radiation source. This fully considers the characteristics of fingerprint features, and then uses a support vector machine classifier to classify and identify the selected statistical features to obtain the category of the communication radiation source. This reduces the overall complexity of the algorithm while ensuring high identification accuracy. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0033] Figure 1 This is a flowchart illustrating the communication radiation source identification method provided by the present invention;

[0034] Figure 2 This is a schematic diagram of the algorithm for the communication radiation source identification method provided by the present invention;

[0035] Figure 3 This is a schematic diagram of the signal and its IMF components in the communication radiation source identification method provided by the present invention;

[0036] Figure 4 This is a schematic diagram of the Hilbert time spectrum of the signal in the communication radiation source identification method provided by the present invention;

[0037] Figure 5 This is a schematic diagram of the Hilbert marginal spectrum of the communication radiation source identification method provided by the present invention;

[0038] Figure 6This is a schematic diagram of the communication radiation source identification device provided by the present invention;

[0039] Figure 7 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0041] The following is combined with Figures 1-7 The present invention describes a communication radiation source identification method, apparatus, electronic device, and medium.

[0042] Figure 1 This is a flowchart illustrating the communication radiation source identification method provided by the present invention. (Refer to...) Figure 1 The communication radiation source identification method provided by the present invention includes steps 110 and 120.

[0043] Step 110: Based on the Hilbert marginal spectrum corresponding to the signal of the radiation source to be identified, determine the statistical characteristics of the Hilbert marginal spectrum. The statistical characteristics are used to indicate the fingerprint characteristics of the communication radiation source. The statistical characteristics include the mean and variance of the Hilbert marginal spectrum.

[0044] Step 120: Input the statistical features into the support vector machine classifier to obtain the radiation source category corresponding to the radiation source signal to be identified.

[0045] The execution subject of the communication radiation source identification method provided by this invention can be an electronic device, a component in an electronic device, an integrated circuit, or a chip. The electronic device can be a mobile electronic device or a non-mobile electronic device. For example, mobile electronic devices can be mobile phones, tablets, laptops, handheld computers, in-vehicle electronic devices, wearable devices, ultra-mobile personal computers (UMPCs), netbooks, or personal digital assistants (PDAs), etc., while non-mobile electronic devices can be servers, network-attached storage (NAS), personal computers (PCs), televisions (TVs), ATMs, or self-service machines, etc. This invention does not impose specific limitations.

[0046] The technical solution of the present invention will be described in detail below using the example of a computer executing the communication radiation source identification method provided by the present invention.

[0047] It should be noted that communication radiation sources refer to electronic devices that generate communication signals, such as mobile phones, wireless routers, and base stations, which are common communication radiation sources.

[0048] Since the output signals of different communication radiation sources have different characteristics in both the time and frequency domains, the Hilbert-Huang Transform (HHT) has high time-frequency resolution and good time-frequency aggregation, and analyzes signal characteristics based on the non-normalized joint weighted distribution of energy, time, and frequency.

[0049] Therefore, the present invention can select HHT to process the output signal of the communication radiation source, and can extract local features of the finite-length signal.

[0050] like Figure 2 As shown, the output signal s(t) of the communication radiation source to be identified is received, then the Hilbert marginal spectrum is obtained using HHT, the fingerprint features of the communication radiation source are extracted using the Hilbert marginal spectrum, and finally, a Support Vector Machine (SVM) classifier is used to classify and identify the extracted features. The above process is explained in detail below:

[0051] In step 110, the radiation source signal to be identified is the radiation source signal that needs to be identified for communication radiation source identification.

[0052] The Hilbert marginal spectrum is based on the calculated Hilbert time spectrum. By integrating the Hilbert time spectrum on the time axis, the relationship between amplitude, time, and frequency is transformed into a relationship between amplitude and frequency. The Hilbert marginal spectrum describes the distribution of amplitude (or energy) on the frequency axis.

[0053] This invention extracts statistical features of the Hilbert marginal spectrum as fingerprint features of communication radiation sources for extraction and individual identification.

[0054] It is understandable that electromagnetic signal data contains not only the fingerprint characteristics of communication radiation sources but also the randomness of symbols. To reduce the impact of symbol randomness, this invention selects the mean of the Hilbert marginal spectrum as one of the fingerprint characteristics of communication radiation sources. Due to frequency aliasing caused by hardware damage coupling, the Hilbert marginal spectrum has different energy distributions at different frequencies, which is also an important manifestation of the fingerprint characteristics of communication radiation sources. Therefore, this invention selects variance as a statistical feature to reflect the dispersion of the Hilbert marginal spectrum.

[0055] To simplify the algorithm, this invention selects an SVM classifier for classification and recognition. Compared with convolutional neural networks and multi-domain fusion features, this greatly reduces the overall computational complexity of the algorithm while maintaining a high recognition accuracy.

[0056] It's important to note that Support Vector Machines (SVMs) are a supervised learning model that can perform classification or regression tasks with a relatively small number of samples. Essentially, they are based on finding an optimal separating hyperplane in the sample space from the training set, such that the sum of the distances from the "support vectors" of two different classes to the hyperplane is maximized.

[0057] In step 120, the statistical features of the Hilbert marginal spectrum extracted from the radiation source signal to be identified are input into the SVM classifier for classification and identification, thereby determining the communication radiation source category corresponding to the radiation source signal to be identified.

[0058] The training set of the SVM classifier mainly consists of the statistical features of the Hilbert marginal spectrum and the radiation source category corresponding to the Hilbert marginal spectrum.

[0059] The communication radiation source identification method provided by this invention determines the statistical characteristics of the Hilbert marginal spectrum corresponding to the signal of the radiation source to be identified, and uses the mean and variance in the statistical characteristics as fingerprint features of the communication radiation source. It fully considers the characteristics of fingerprint features, and then uses a support vector machine classifier to classify and identify the selected statistical features to obtain the category of the communication radiation source. This reduces the overall complexity of the algorithm while ensuring high identification accuracy.

[0060] In some embodiments, the support vector machine classifier is obtained as follows:

[0061] Obtain the Hilbert marginal spectrum samples corresponding to the radiation source signal samples;

[0062] Determine the mean and variance of the Hilbert marginal spectrum sample;

[0063] A support vector machine classifier is trained based on the radiation source category corresponding to the radiation source signal sample and the mean and variance of the Hilbert marginal spectrum sample.

[0064] The classifiers used in the communication radiation source identification algorithm provided by this invention include, but are not limited to, support vector machine classifiers, and are not specifically limited here.

[0065] It is understandable that radiation source signal samples are obtained before training the support vector machine classifier. This invention does not impose specific limitations on the method of obtaining radiation source signals.

[0066] Then, the Hilbert marginal spectrum sample corresponding to the radiation source signal sample is determined by HHT.

[0067] Based on the Hilbert marginal spectrum samples, the mean and variance of the Hilbert marginal spectrum samples are determined, as well as the radiation source category corresponding to the radiation source signal samples.

[0068] By randomly selecting a certain proportion of the above data as the training set, the entire dataset can be used as the test set. The training set is first input into the Support Vector Machine (SVM) classifier, and after training, a trained SVM classifier is obtained.

[0069] Understandably, based on the radiation source category corresponding to the radiation source signal sample, the trained support vector machine classifier can be a binary classification model or a multi-class classification model.

[0070] The communication radiation source identification method provided by this invention can improve the classification and identification accuracy by training a support vector machine classifier.

[0071] In some embodiments, training a support vector machine classifier includes:

[0072] We optimize the support vector machine classifier based on the 5-fold cross-validation method.

[0073] It should be noted that the basic idea of ​​Support Vector Machines (SVMs) is to transform the original data into a high-dimensional space and maximize the margin between the two classes of data by finding the optimal hyperplane. This mapping requires the selection of a kernel function.

[0074] To optimize the hyperparameters of the kernel function and achieve higher classification accuracy, this invention employs a random search approach and determines the optimal kernel function and other relevant parameters during training through k-fold cross-validation.

[0075] k-fold cross-validation refers to first randomly dividing the dataset D into k mutually exclusive subsets of equal size. That is, in each round, k-1 subsets are randomly selected as the training set, and the remaining subset is used as the test set. After this round, k subsets are randomly selected again for training. After several rounds (less than k), a loss function is chosen to evaluate the optimal model and parameters. Note that the stability and fidelity of the cross-validation evaluation results largely depend on the value of k.

[0076] In this invention, k = 5, that is, 5-fold cross-validation is used to optimize the support vector machine classifier.

[0077] In some embodiments, prior to step 110, the communication radiation source identification method further includes:

[0078] Acquire the signal of the radiation source to be identified;

[0079] Perform Hilbert-Huang transform on the signal of the radiation source to be identified to obtain the Hilbert time spectrum corresponding to the signal of the radiation source to be identified;

[0080] Based on the Hilbert time spectrum corresponding to the radiation source signal to be identified, the Hilbert marginal spectrum corresponding to the radiation source signal to be identified is determined.

[0081] It should be noted that HHT mainly consists of two steps: Empirical Mode Decomposition (EMD) and Hilbert Transform.

[0082] The basic process of HHT is as follows: First, the given original signal is decomposed into several Intrinsic Mode Functions (IMFs) using EMD:

[0083]

[0084] Where s(t) is the original signal, imfi i (t) represents the intrinsic mode functions of each layer, r N (t) is the original signal s(t) decomposed into the intrinsic mode functions imf of each layer. i The remaining component after (t).

[0085] Then perform the Hilbert-Huang transform on each IMF:

[0086]

[0087] The Hilbert time spectrum for each IMF is obtained as follows:

[0088]

[0089] Among them, a i (t) is the instantaneous amplitude of the IMF component, ω i (t) represents the instantaneous frequency of the IMF component.

[0090] a i (t) and ω i (t) respectively satisfy:

[0091]

[0092]

[0093] Finally, by combining the Hilbert time spectra of all IMFs, we can obtain the Hilbert time spectrum of the original signal:

[0094]

[0095] like Figure 3 As shown, the signal of the radiation source to be identified, its IMF components obtained after EMD decomposition, and the residuals are presented. IMF1 has the highest frequency and the smallest characteristic time scale. As the number of decomposition layers increases, the characteristic time scale of the IMF components increases, and the frequency decreases. Different frequency components in the multi-component signal are separated. After obtaining the IMF components, a Hilbert transform is performed on them to construct an analytic signal, obtain the instantaneous frequency, and thus obtain the final total Hilbert time spectrum of the signal, as shown below. Figure 4 As shown.

[0096] from Figure 4 As can be seen, the frequency components corresponding to different times of the signal can be clearly displayed, and the energy distribution in different time and frequency domains can be judged by the display color.

[0097] Combined with coupling analysis, aliasing and harmonics are the main manifestations of fingerprint features, and they are more pronounced in the signal frequency domain. Therefore, to extract amplitude feature information in the frequency domain, the Hilbert marginal spectrum is used to further extract fingerprint features based on the Hilbert time spectrum after HHT transformation. The Hilbert marginal spectrum is obtained by integrating the Hilbert time spectrum on the time axis:

[0098]

[0099] As can be seen from formulas (3)-(6), the Hilbert marginal spectrum reflects the change of signal amplitude with frequency, representing the cumulative amplitude that is statistically significant over the entire signal time span, such as... Figure 5 As shown.

[0100] Compared to the Fourier transform, the Hilbert marginal spectrum breaks free from the constraint that a signal requires a complete period of waveform. This not only improves the efficiency of signal representation but also allows for the representation of variable frequencies, enabling more accurate local analysis of signals and the extraction of more subtle waveform change features. As a result, the fingerprint features of communication radiation sources can be amplified.

[0101] In some embodiments, after training the support vector machine classifier, the communication radiation source identification method further includes:

[0102] Based on the hardware damage parameters of the communication radiation source corresponding to the statistical characteristics, a sensitivity analysis is performed on the trained support vector machine classifier to determine the sensitive parameters that affect the classification and recognition performance.

[0103] The hardware impairment parameters include: filter distortion parameters, IQ imbalance gain imbalance parameters, IQ imbalance phase imbalance parameters, and different order coefficients of the power amplifier; the sensitive parameters include: IQ imbalance gain imbalance parameters, first-order coefficient of power amplifier nonlinear error, and third-order coefficient of power amplifier nonlinear error.

[0104] To explore the applicability of the communication radiation source identification method provided by this invention, sensitivity analysis algorithms can be used to perform sensitivity analysis on the classifier used in the communication radiation source identification method, such as the variance-based global sensitivity analysis method Sobol algorithm, to distinguish the independent and interactive sensitivities of hardware damage parameters that introduce fingerprint features.

[0105] For example, the parameters corresponding to the three types of hardware damage parameters introduced include:

[0106] Filter distortion parameters ρ1 and The gain imbalance parameter α and phase imbalance parameter ε of IQ imbalance, the coefficients a1, a3 and a5 of different order terms of the power amplifier, are used to generate a dataset by Monte Carlo sampling within a specified parameter range using a low-difference SOBO sequence, and the selected algorithm is used to classify and identify communication radiation sources.

[0107] Here, IQ stands for In-phase and Quadrature, meaning in-phase and quadrature.

[0108] Based on the classification and identification results, sensitivity evaluation indicators such as the first-order influence index and the total effect index are calculated to assess performance.

[0109] In the three hardware impairment coupling scenarios illustrated, the identification performance is highly sensitive to the gain imbalance parameter α of IQ imbalance and the first-order coefficient a1 and third-order coefficient a3 of power amplifier nonlinear error.

[0110] The communication radiation source identification device provided by the present invention is described below. The communication radiation source identification device described below can be referred to in correspondence with the communication radiation source identification method described above.

[0111] Figure 6 This is a schematic diagram of the communication radiation source identification device provided by the present invention. (Refer to...) Figure 6 The communication radiation source identification device provided by the present invention includes: a determination module 610 and a classification module 620.

[0112] The determination module 610 is used to determine the statistical characteristics of the Hilbert marginal spectrum based on the Hilbert marginal spectrum corresponding to the signal of the radiation source to be identified. The statistical characteristics are used to indicate the fingerprint characteristics of the communication radiation source. The statistical characteristics include the mean and variance of the Hilbert marginal spectrum.

[0113] The classification module 620 is used to input the statistical features into a support vector machine classifier to obtain the radiation source category corresponding to the radiation source signal to be identified.

[0114] The communication radiation source identification device provided by this invention determines the statistical characteristics of the Hilbert marginal spectrum corresponding to the signal of the radiation source to be identified, and uses the mean and variance in the statistical characteristics as fingerprint features of the communication radiation source. It fully considers the characteristics of fingerprint features, and then uses a support vector machine classifier to classify and identify the selected statistical features to obtain the category of the communication radiation source. This reduces the overall complexity of the algorithm while ensuring high identification accuracy.

[0115] In some embodiments, the apparatus further includes:

[0116] The training module is used to obtain Hilbert marginal spectrum samples corresponding to radiation source signal samples;

[0117] Determine the mean and variance of the Hilbert marginal spectrum sample;

[0118] The support vector machine classifier is trained based on the radiation source category corresponding to the radiation source signal sample, and the mean and variance of the Hilbert marginal spectrum sample.

[0119] In some embodiments, the apparatus further includes:

[0120] The transformation module is used to acquire the radiation source signal to be identified before determining the statistical characteristics of the Hilbert marginal spectrum based on the Hilbert marginal spectrum corresponding to the radiation source signal to be identified;

[0121] Perform a Hilbert-Huang transform on the radiation source signal to be identified to obtain the Hilbert time spectrum corresponding to the radiation source signal to be identified;

[0122] Based on the Hilbert time spectrum corresponding to the radiation source signal to be identified, the Hilbert marginal spectrum corresponding to the radiation source signal to be identified is determined.

[0123] In some embodiments, the apparatus further includes:

[0124] The analysis module is used to perform sensitivity analysis on the trained support vector machine classifier based on the hardware damage parameters of the communication radiation source corresponding to the statistical features after the support vector machine classifier is trained, and to determine the sensitive parameters that affect the classification and recognition performance.

[0125] The hardware impairment parameters include: filter distortion parameters, IQ imbalance gain imbalance parameters, IQ imbalance phase imbalance parameters, and different order coefficients of the power amplifier; the sensitive parameters include: IQ imbalance gain imbalance parameters, first-order coefficient of power amplifier nonlinear error, and third-order coefficient of power amplifier nonlinear error.

[0126] In some embodiments, the training module is specifically used for:

[0127] The support vector machine classifier is optimized based on the 5-fold cross-validation method.

[0128] Figure 7 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 7 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730, and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other via the communication bus 740. The processor 710 can call logical instructions in the memory 730 to execute a communication radiation source identification method, which includes:

[0129] Based on the Hilbert marginal spectrum corresponding to the signal of the radiation source to be identified, the statistical characteristics of the Hilbert marginal spectrum are determined. The statistical characteristics are used to indicate the fingerprint characteristics of the communication radiation source. The statistical characteristics include the mean and variance of the Hilbert marginal spectrum.

[0130] The statistical features are input into a support vector machine classifier to obtain the radiation source category corresponding to the radiation source signal to be identified.

[0131] Furthermore, the logical instructions in the aforementioned memory 730 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0132] On the other hand, the present invention also provides a computer program product, the computer program product comprising a computer program that can be stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer is able to execute the communication radiation source identification method provided by the above methods, the method comprising:

[0133] Based on the Hilbert marginal spectrum corresponding to the signal of the radiation source to be identified, the statistical characteristics of the Hilbert marginal spectrum are determined. The statistical characteristics are used to indicate the fingerprint characteristics of the communication radiation source. The statistical characteristics include the mean and variance of the Hilbert marginal spectrum.

[0134] The statistical features are input into a support vector machine classifier to obtain the radiation source category corresponding to the radiation source signal to be identified.

[0135] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to perform the communication radiation source identification method provided by the above methods, the method comprising:

[0136] Based on the Hilbert marginal spectrum corresponding to the signal of the radiation source to be identified, the statistical characteristics of the Hilbert marginal spectrum are determined. The statistical characteristics are used to indicate the fingerprint characteristics of the communication radiation source. The statistical characteristics include the mean and variance of the Hilbert marginal spectrum.

[0137] The statistical features are input into a support vector machine classifier to obtain the radiation source category corresponding to the radiation source signal to be identified.

[0138] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0139] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method of communication emitter identification, the method comprising: receiving a signal from a communication emitter; determining a signal characteristic of the signal; and identifying the communication emitter based on the signal characteristic. include: Based on the Hilbert marginal spectrum corresponding to the signal of the radiation source to be identified, the statistical characteristics of the Hilbert marginal spectrum are determined. The statistical characteristics are used to indicate the fingerprint characteristics of the communication radiation source. The statistical characteristics include the mean and variance of the Hilbert marginal spectrum. The Hilbert marginal spectrum describes the distribution of amplitude or energy on the frequency axis. The variance of the Hilbert marginal spectrum characterizes the degree of dispersion of the Hilbert marginal spectrum. The statistical features are input into a support vector machine classifier to obtain the radiation source category corresponding to the radiation source signal to be identified. The method further includes: Based on the hardware damage parameters of the communication radiation source corresponding to the statistical characteristics, a sensitivity analysis is performed on the trained support vector machine classifier to determine the sensitive parameters that affect the classification and recognition performance. The hardware impairment parameters include: filter distortion parameters, IQ imbalance gain imbalance parameters, IQ imbalance phase imbalance parameters, and different order coefficients of the power amplifier; the sensitive parameters include: IQ imbalance gain imbalance parameters, first-order coefficient of power amplifier nonlinear error, and third-order coefficient of power amplifier nonlinear error.

2. The method of claim 1, wherein, The support vector machine classifier is obtained in the following way: Obtain the Hilbert marginal spectrum samples corresponding to the radiation source signal samples; Determine the mean and variance of the Hilbert marginal spectrum sample; The support vector machine classifier is trained based on the radiation source category corresponding to the radiation source signal sample, and the mean and variance of the Hilbert marginal spectrum sample.

3. The method of claim 1, wherein, Before determining the statistical characteristics of the Hilbert marginal spectrum based on the Hilbert marginal spectrum corresponding to the radiation source signal to be identified, the method further includes: Acquire the signal of the radiation source to be identified; Perform a Hilbert-Huang transform on the radiation source signal to be identified to obtain the Hilbert time spectrum corresponding to the radiation source signal to be identified; Based on the Hilbert time spectrum corresponding to the radiation source signal to be identified, the Hilbert marginal spectrum corresponding to the radiation source signal to be identified is determined.

4. The method of claim 2, wherein, Training the support vector machine classifier includes: The support vector machine classifier is optimized based on the 5-fold cross-validation method.

5. A communication radiation source identification apparatus based on the communication radiation source identification method according to any one of claims 1 to 4, characterized by include: The determination module is used to determine the statistical characteristics of the Hilbert marginal spectrum based on the Hilbert marginal spectrum corresponding to the signal of the radiation source to be identified. The statistical characteristics are used to indicate the fingerprint characteristics of the communication radiation source. The statistical characteristics include the mean and variance of the Hilbert marginal spectrum. The classification module is used to input the statistical features into a support vector machine classifier to obtain the radiation source category corresponding to the radiation source signal to be identified.

6. The apparatus of claim 5, wherein, The device further includes: The training module is used to obtain Hilbert marginal spectrum samples corresponding to radiation source signal samples; Determine the mean and variance of the Hilbert marginal spectrum sample; The support vector machine classifier is trained based on the radiation source category corresponding to the radiation source signal sample, and the mean and variance of the Hilbert marginal spectrum sample.

7. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the communication radiation source identification method as described in any one of claims 1 to 4.

8. A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by the processor, it implements the communication radiation source identification method as described in any one of claims 1 to 4.

9. A computer program product comprising a computer program, characterized in that, The computer program, when executed by a processor, implements the communication radiation source identification method according to any one of claims 1 to 4.

Images