Multi-modulation signal blind detection and identification method, device, computer equipment and medium

Abstract

The present invention discloses a method, apparatus, computer equipment, and medium for blind detection and identification of multiple modulated signals. The method comprises the following steps: inputting a quasi-baseband signal of a modulation signal to be detected and calculating the fourth power spectrum of the quasi-baseband signal, wherein the modulation signal to be detected is one of GMSK, FQPSK, and OQPSK modulation signals; extracting the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude from the fourth power spectrum; and determining the type of the modulation signal to be detected based on the magnitude relationship between the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude. The present invention can identify GMSK, FQPSK, and OQPSK modulation signals and has the advantages of simple implementation, low cost, high recognition efficiency and accuracy, low resource consumption, and strong robustness.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless communication signal recognition, and in particular to a method, device, computer equipment and medium for blind detection and recognition of multi-modulation signals. Background Art

[0002] In wireless communication signal identification and monitoring systems, signal classification is the final and most crucial step in the overall system. GMSK (Gaussian Filtered Minimum Shift Keying) modulation can be simply described as a partial response CPM modulation scheme with a modulation index of 0.5. It is achieved by filtering the rectangular frequency pulses of the MSK signal using a filter with a Gaussian impulse response before carrier frequency modulation. OQPSK (offset-QPSK) modulation is a constant envelope data modulation technique developed based on QPSK (Quadrature Phase Shift Keying). It is an improved version of QPSK modulation, also known as offset-QPSK modulation. It shares the same phase relationship as QPSK modulation, splitting the input code stream into two paths and then performing quadrature modulation. The difference is that the in-phase and quadrature branches are temporally offset by half a symbol period. Therefore, the OQPSK signal does not experience phase flips exceeding 90°, and the time-domain envelope variation is smaller. FQPSK modulation is a further improvement of OQPSK modulation. It eliminates the 3dB envelope fluctuation by introducing cross-correlation, thereby making the signal envelope constant in the time domain.

[0003] Since GMSK, FQPSK and OQPSK are all quasi-constant envelope or constant envelope signals, it is difficult to extract effective features in the time domain, especially under low signal-to-noise ratio. After performing nonlinear transformation operations on the signal, the signal may have discrete spectral line features in the frequency domain, such as Figures 1 to 3 As shown in the figure, the spectra and high-order characteristics of GMSK, FQPSK, and OQPSK modulation schemes are very similar, both in terms of the number of spectral peaks and their relative positions. Therefore, without prior knowledge of the precise signal bandwidth and symbol rate, it is difficult to effectively distinguish between these three modulation schemes. Although the spectral peak amplitudes of different signals vary, these peak amplitudes are significantly affected by the signal-to-noise ratio and exhibit low robustness. Precisely because GMSK, FQPSK, and OQPSK modulated signals all share similar basic high-order spectral line characteristics and high-order cumulants, current signal system identification algorithms typically classify GMSK, FQPSK, and OQPSK modulated signals as a single subclass, making further differentiation difficult. Summary of the Invention

[0004] The technical problem to be solved by the present invention is: in response to the technical problems existing in the prior art, the present invention provides a multi-modulation signal blind detection and identification method, device, computer equipment and medium with simple implementation method, low cost, high recognition efficiency and accuracy, low resource consumption and strong robustness, which can accurately and quickly realize the identification of GMSK, FQPSK and OQPSK modulated signals.

[0005] In order to solve the above technical problems, the technical solution proposed by the present invention is:

[0006] A method for blind detection and identification of multi-modulation signals, comprising the following steps:

[0007] Input a quasi-baseband signal of a modulation signal to be detected and calculate the fourth power spectrum of the quasi-baseband signal, wherein the type of the modulation signal to be detected is one of GMSK, FQPSK and OQPSK modulation signals;

[0008] Extracting the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude from the fourth power spectrum;

[0009] The type of the modulation signal to be detected is determined according to the magnitude relationship between the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude.

[0010] Furthermore, the quartic spectrum is obtained by first performing a quartic nonlinear transformation on the input signal and then calculating the power spectrum.

[0011] Furthermore, before extracting the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude from the quartic spectrum, the method further includes performing spectral line highlighting processing on the quartic spectrum to highlight the local maximum value, thereby obtaining a processed quartic spectrum.

[0012] Furthermore, extracting the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude from the quartic spectrum includes:

[0013] Extracting the peak amplitude of the spectrum from the current upper fourth power spectrum according to the peak position to obtain the maximum peak amplitude;

[0014] After setting the amplitudes of the maximum peak position in the current quartic spectrum and the specified positions to the left and right of the maximum peak position to zero, the maximum peak position in the current quartic spectrum is searched again and the corresponding peak amplitude is extracted to obtain the second maximum peak amplitude.

[0015] Furthermore, by calculating the ratio Ra between the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude, the type of the modulation signal to be detected is determined according to the size of the ratio Ra.

[0016] Furthermore, if the ratio Ra is greater than a preset first threshold, the modulation signal to be detected is determined to be an OQPSK modulation signal; if the ratio Ra is less than the preset first threshold and greater than the preset second threshold, the modulation signal to be detected is determined to be an FQPSK modulation signal; if the ratio Ra is less than the preset second threshold, the modulation signal to be detected is determined to be a GMSK modulation signal, and the preset second threshold is less than the preset first threshold.

[0017] Furthermore, the preset first threshold is 10, and the preset second threshold is 2.

[0018] A multi-modulation signal blind detection and identification device, comprising:

[0019] a spectrum calculation module, configured to input a quasi-baseband signal of a modulation signal to be detected and calculate a fourth power spectrum of the quasi-baseband signal, wherein the modulation signal to be detected is one of GMSK, FQPSK and OQPSK modulation signals;

[0020] A parameter extraction module, configured to extract a maximum spectrum peak amplitude and a second maximum spectrum peak amplitude from the quartic spectrum;

[0021] The signal type determination module is configured to determine the type of the modulation signal to be detected according to the magnitude relationship between the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude.

[0022] A computer device includes a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to execute the computer program to perform the above method.

[0023] A computer-readable storage medium storing a computer program, wherein the computer program implements the above method when executed.

[0024] Compared with the existing technology, the advantages of the present invention are: the present invention extracts the fourth power spectrum of the signal to be detected (one of the GMSK, OQPSK and FQPSK modulated signals), extracts the maximum spectral line and the second maximum spectral line amplitudes, and uses the size relationship between the maximum spectral line and the second maximum spectral line amplitudes as characteristic parameters to obtain classification characteristic parameters that do not require prior information and are insensitive to the signal-to-noise ratio. Based on the characteristic parameters, GMSK, OQPSK and FQPSK modulated signals can be accurately and quickly identified without prior information, while ensuring the robustness of the identification. Moreover, since only the positions and amplitudes of the maximum spectral peak and the second maximum spectral peak need to be searched, very precise parameters such as carrier frequency and signal bandwidth are not required, and the complexity of implementation can be greatly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a schematic diagram of the spectrum and high-order spectrum of the MSK modulated signal.

[0026] Figure 2 This is a schematic diagram of the spectrum and high-order spectrum of the OQPSK modulated signal.

[0027] Figure 3 This is a schematic diagram of the spectrum and high-order spectrum of the FQPSK modulated signal.

[0028] Figure 4 It is the fourth power spectrum of GMSK, FQPSK and OQPSK signals.

[0029] Figure 5 This is a schematic diagram of the implementation flow of the multi-modulation signal blind detection and identification method of this embodiment.

[0030] Figure 6 It is a schematic diagram of the effects before and after the spectral line highlighting process in a specific application embodiment.

[0031] Figure 7 FIG. 1 is a schematic diagram of Ra simulation results of GMSK, FQPSK and OQPSK signals in a specific application embodiment.

[0032] Figure 8 The figure is a flow chart of using the ratio Ra to realize GMSK, FQPSK and OQPSK signal identification in a specific application embodiment. DETAILED DESCRIPTION

[0033] The present invention will be further described below in conjunction with the accompanying drawings and specific preferred embodiments, but the scope of protection of the present invention is not limited thereby.

[0034] Taking into account that the number of spectral lines, their positions and intensities in the spectra of GMSK, OQPSK and FQPSK modulation modes are extremely similar, making them difficult to distinguish, the present invention first conducts an in-depth analysis of the fourth power spectra of the three modulation modes of GMSK, OQPSK and FQPSK, and finds that the fourth power spectra of GMSK, FQPSK and OQPSK signals have the following characteristics: there are multiple discrete spectral lines, and the spectral line amplitudes in the fourth power spectra corresponding to different modulation modes will be significantly different, that is, there will be a relatively obvious difference in the size relationship between the maximum spectral peak amplitude and the second largest spectral peak amplitude.

[0035] Taking the baseband signal with 8 times oversampling as an example, Figure 4 (a), (b) and (c) correspond to the fourth power spectra of GMSK, FQPSK and OQPSK signals respectively. Figure 4 (d) corresponds to the local details of OQPSK. Figure 4As can be seen from the figure, each quartic spectrum has a common characteristic: the presence of three discrete spectral lines: a maximum spectral line at zero frequency (i.e., a quadruple spectral line; due to down-conversion, the signal carrier frequency is already zero frequency) and a bilaterally symmetrical second-maximum spectral line. The amplitudes of the maximum and second-maximum spectral lines corresponding to different modulation schemes differ significantly. The ratio between the maximum and second-maximum spectral lines in the quartic spectrum of an OQPSK signal is large, while the ratio between the maximum and second-maximum spectral lines in the quartic spectrum of a GMSK signal is small. The ratio between the maximum and second-maximum spectral lines in the quartic spectrum of an FQPSK signal is smaller than that of an OQPSK signal and larger than that of a GMSK signal. Therefore, the relationship between the spectral line amplitudes in the quartic spectrum can be used to identify the corresponding modulation system type.

[0036] The present invention utilizes the characteristic that the spectral peak amplitudes in the fourth power spectrum of GMSK, FQPSK and OQPSK signals are significantly different. By extracting the fourth power spectrum of the signal to be detected (one of the GMSK, OQPSK and FQPSK modulated signals), the maximum spectral line and the second maximum spectral line amplitudes are extracted, and then a classification feature parameter is constructed based on the size relationship between the maximum spectral line and the second maximum spectral line amplitude. The parameter does not require prior information, has high robustness, has obvious differences, and is insensitive to the signal-to-noise ratio. Based on the feature parameter, GMSK, OQPSK and FQPSK modulated signals can be accurately and quickly identified without prior information, while also ensuring the robustness of the identification.

[0037] like Figure 5 As shown, the steps of the multi-modulation signal blind detection and identification method of this embodiment include:

[0038] S01. Spectrum calculation: Input the quasi-baseband signal of the modulation signal to be detected and calculate the fourth power spectrum of the quasi-baseband signal. The modulation signal to be detected is one of GMSK, FQPSK and OQPSK modulation signals.

[0039] In this embodiment, the input modulated signal to be detected is specifically a quasi-baseband signal that has undergone down-conversion and decimation filtering (a certain frequency deviation and oversampling greater than 4 times are allowed). By first performing a fourth-power nonlinear transformation on the input signal and then calculating the power spectrum, the fourth-power spectrum of the input signal, which is also the original spectrum, is obtained.

[0040] S02. Parameter extraction: extract the maximum spectral peak amplitude and the second maximum spectral peak amplitude from the quartic spectrum.

[0041] This embodiment further includes performing line highlighting on the quartic spectrum to highlight the local maximum value before extracting the maximum peak amplitude and the second maximum peak amplitude from the quartic spectrum, thereby obtaining a processed quartic spectrum. Specifically, the line highlighting process of the quartic spectrum can be performed according to the following formula:

[0042]

[0043] in, is the processed quartic spectrum, S(k) is the original spectrum, and S(i) is the amplitude of the i-th spectrum in the original spectrum.

[0044] Taking the fourth power spectrum of OQPSK signal as an example, the spectrum line highlighting process before and after is performed in a specific application embodiment. Figure 6 As shown, Figure 6 (a) corresponds to the original spectrum, and (b) corresponds to the spectrum after the spectral lines are highlighted. Figure 6 It can be seen that after the spectral line highlighting processing, the local maximum value in the original spectrum can be effectively highlighted.

[0045] In this embodiment, the specific steps of extracting the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude from the quartic spectrum include:

[0046] S201. Extract the peak amplitude of the spectrum according to the peak position in the current fourth power spectrum to obtain the maximum peak amplitude Q1;

[0047] S202. After setting the amplitudes of the maximum peak position in the current quartic spectrum and the specified number of positions to the left and right of the maximum peak position to zero, re-search the maximum peak position in the current quartic spectrum and extract the corresponding peak amplitude to obtain the next maximum peak amplitude Q2.

[0048] Specifically, this embodiment first searches for the maximum spectral peak position on the fourth power spectrum after the spectral line highlighting processing, and extracts the spectral peak amplitude on the original spectrum according to the spectral peak position, that is, obtains the maximum spectral peak amplitude Q1; then sets the maximum spectral peak position and the amplitudes of the four points on the left and right of the fourth power spectrum after the spectral line highlighting processing to zero, and then searches for the maximum spectral peak position again, and extracts the spectral peak amplitude on the spectrum according to the spectral peak position, that is, obtains the second largest spectral peak amplitude Q2.

[0049] It is understandable that the maximum spectrum peak amplitude Q1 and the second maximum spectrum peak amplitude Q2 can of course be extracted using other extraction methods according to actual needs.

[0050] S03. Signal type determination: Determine the type of the modulation signal to be detected based on the magnitude relationship between the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude.

[0051] Considering that if the spectrum line amplitude is simply taken as the characteristic parameter, it will be easily affected by the signal power and signal-to-noise ratio, and it is impossible to give a suitable threshold to judge and distinguish, and the ratio Ra between the maximum spectrum line Q1 and the second maximum spectrum line amplitude Q2 corresponding to different modulation modes will have obvious differences. For example, the ratio between the maximum spectrum line and the second maximum spectrum line amplitude in the fourth power spectrum of the OQPSK signal is large, while the ratio between the maximum spectrum line and the second maximum spectrum line amplitude in the fourth power spectrum of the GMSK signal is small, and the ratio between the maximum spectrum line and the second maximum spectrum line amplitude in the fourth power spectrum of the FQPSK signal is small. The ratio between the line amplitudes is smaller than that of the OQPSK signal and larger than that of the GMSK signal. Since the ratio Ra is the ratio of the maximum value to the second maximum value, it represents the inherent characteristics of the signal itself and does not change with the signal power or signal-to-noise ratio. In this embodiment, the ratio Ra between the maximum spectrum peak amplitude Q1 and the second maximum spectrum peak amplitude Q2 is calculated as a characteristic parameter. The calculation expression is shown in formula (2). The type of the modulation signal to be detected is judged according to the size of the ratio Ra, thereby realizing the recognition of GMSK, FQPSK and OQPSK modulation signals.

[0052]

[0053] Among them, Q1 is the maximum value of the fourth power spectrum peak amplitude; Q2 is the second maximum value of the fourth power spectrum peak amplitude.

[0054] In a specific application embodiment, the symbol rate is 1000sps and 8 times oversampling is used. After 100 Monte Carlo simulations, the Ra simulation results of the three modulation modes of GMSK, FQPSK and OQPSK are obtained as follows: Figure 7 As shown, Figure 7 (a) corresponds to the Ra simulation result, and (b) corresponds to the local detail diagram. Figure 7 As can be seen, the ratio Ra is very stable under different signal-to-noise ratios, reflecting the inherent modulation characteristics of the signal itself. An EbN0 above 5 dB can effectively distinguish the three modulation modes. This means that by using the ratio Ra as a characteristic parameter, this embodiment can effectively identify GMSK, FQPSK, and OQPSK modulation signals without requiring prior information and without being sensitive to signal-to-noise ratio. Furthermore, since only the positions and amplitudes of the maximum and second-largest spectral peaks need to be searched, precise parameters such as carrier frequency and signal bandwidth are not required. This significantly reduces implementation complexity and is highly beneficial for engineering implementation.

[0055] like Figure 8As shown, in this embodiment, if the ratio Ra is greater than a preset first threshold, the modulation signal to be detected is determined to be an OQPSK modulation signal. If the ratio Ra is less than the preset first threshold and greater than a preset second threshold, the modulation signal to be detected is determined to be an FQPSK modulation signal. If the ratio Ra is less than the preset second threshold, the modulation signal to be detected is determined to be a GMSK modulation signal, and the preset second threshold is less than the preset first threshold. The preset first threshold is 10, and the preset second threshold is 2. The specific values ​​of the preset first and second thresholds can be adjusted based on actual conditions.

[0056] It can be understood that in addition to using the above-mentioned ratio Ra as a classification feature parameter, other types of parameters that can characterize the size relationship between the maximum spectral peak and the second largest spectral peak amplitude in the three modulation signals of GMSK, FQPSK and OQPSK can also be used according to actual needs to realize the identification of the three modulation signals.

[0057] The multi-modulation signal blind detection and identification device of this embodiment includes:

[0058] a spectrum calculation module, configured to input a quasi-baseband signal of a modulation signal to be detected and calculate a fourth power spectrum of the quasi-baseband signal, wherein the modulation signal to be detected is one of GMSK, FQPSK and OQPSK modulation signals;

[0059] A parameter extraction module, configured to extract a maximum spectrum peak amplitude and a second maximum spectrum peak amplitude from the quartic spectrum;

[0060] The signal type determination module is configured to determine the type of the modulation signal to be detected according to the magnitude relationship between the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude.

[0061] In this embodiment, the parameter extraction module includes:

[0062] The first extraction unit is used to extract the spectrum peak amplitude from the current upper fourth spectrum according to the spectrum peak position to obtain the maximum spectrum peak amplitude;

[0063] The second extraction unit is used to set the amplitudes of the maximum spectral peak position and the specified number of positions to the left and right of the maximum spectral peak position in the current quartic spectrum to zero, then re-search the maximum spectral peak position in the current quartic spectrum and extract the corresponding spectral peak amplitude to obtain the second maximum spectral peak amplitude.

[0064] In this embodiment, the signal type determination module specifically calculates the ratio Ra between the maximum spectral peak amplitude and the second-largest spectral peak amplitude, and determines the type of the modulation signal to be detected based on the magnitude of the ratio Ra. If the ratio Ra is greater than a preset first threshold, the modulation signal to be detected is determined to be an OQPSK modulation signal; if the ratio Ra is less than the preset first threshold and greater than a preset second threshold, the modulation signal to be detected is determined to be an FQPSK modulation signal; and if the ratio Ra is less than a preset second threshold, the modulation signal to be detected is determined to be a GMSK modulation signal, where the preset second threshold is less than the preset first threshold.

[0065] The multi-modulation signal blind detection and identification device of this embodiment corresponds one-to-one to the above-mentioned multi-modulation signal blind detection and identification method, and will not be described in detail here.

[0066] This embodiment further provides a computer device, including a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to execute the computer program to perform the above method.

[0067] This embodiment also provides a computer-readable storage medium storing a computer program, which implements the above method when executed.

[0068] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed above with reference to the preferred embodiment, it is not intended to limit the present invention. Therefore, any simple modifications, equivalent variations, and modifications to the above embodiment that do not depart from the technical solution of the present invention and are based on the technical essence of the present invention shall fall within the scope of protection of the technical solution of the present invention.

Claims

1. A method for blind detection and identification of multi-modulation signals, characterized in that the steps include: Input a quasi-baseband signal of a modulation signal to be detected and calculate the fourth power spectrum of the quasi-baseband signal, wherein the type of the modulation signal to be detected is one of GMSK, FQPSK and OQPSK modulation signals; Extracting the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude from the fourth power spectrum; Calculate the ratio between the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude Ra , according to the ratio Ra The type of modulation signal to be detected is determined by the size of .

2. The method for blind detection and identification of multi-modulation signals according to claim 1, wherein: The quartic spectrum is obtained by first performing a quartic nonlinear transformation on the input signal and then calculating the power spectrum.

3. The method for blind detection and identification of multi-modulation signals according to claim 1, wherein: Before extracting the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude from the quartic spectrum, the method further includes performing spectral line highlighting processing on the quartic spectrum to highlight the local maximum value, thereby obtaining a processed quartic spectrum.

4. The method for blind detection and identification of multi-modulation signals according to claim 1, wherein: The extracting the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude from the quartic spectrum includes: Extracting the peak amplitude of the spectrum from the current upper fourth power spectrum according to the peak position to obtain the maximum peak amplitude; After setting the amplitudes of the maximum peak position in the current quartic spectrum and the specified positions to the left and right of the maximum peak position to zero, the maximum peak position in the current quartic spectrum is searched again and the corresponding peak amplitude is extracted to obtain the second maximum peak amplitude.

5. The method for blind detection and identification of multi-modulation signals according to claim 1, wherein: If the ratio Ra is greater than a preset first threshold, it is determined that the modulation signal to be detected is an OQPSK modulation signal. If the ratio Ra is less than a preset first threshold and greater than a preset second threshold, it is determined that the modulation signal to be detected is an FQPSK modulation signal. If the ratio Ra If the value is less than a preset second threshold, it is determined that the modulation signal to be detected is a GMSK modulation signal, and the preset second threshold is less than the preset first threshold.

6. The method for blind detection and identification of multi-modulation signals according to claim 5, characterized in that: The preset first threshold is 10, and the preset second threshold is 2.

7. A multi-modulation signal blind detection and identification device, characterized in that: include: a spectrum calculation module, configured to input a quasi-baseband signal of a modulation signal to be detected and calculate a fourth power spectrum of the quasi-baseband signal, wherein the modulation signal to be detected is one of GMSK, FQPSK and OQPSK modulation signals; A parameter extraction module, configured to extract a maximum spectrum peak amplitude and a second maximum spectrum peak amplitude from the quartic spectrum; A signal type discrimination module is used to calculate the ratio between the maximum spectrum peak amplitude and the second maximum spectrum peak amplitude. Ra , according to the ratio Ra The type of modulation signal to be detected is determined by the size of .

8. A computer device comprising a processor and a memory, wherein the memory is used to store a computer program, wherein: The processor is configured to execute the computer program to perform the method according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed, the method according to any one of claims 1 to 6 is implemented.

Images