A method, system, device and medium for identifying a disturbed state of a PSK digitally modulated communication signal

Abstract

A method, system, device and medium for identifying the disturbed state of a PSK digital modulation communication signal, the method comprising: first, analyzing the characteristic parameters of signals of different modulation types, including the frequency spectrum symmetry, the number of frequency spectrum discrete peaks, the time domain amplitude entropy, the power spectrum entropy and the high-order cumulative quantity ratio characteristic parameters, to obtain the characteristic parameters for disturbed state identification; second, determining the stable range of PSK characteristic parameters without interference according to the frequency spectrum symmetry, the number of frequency spectrum discrete peaks, the time domain amplitude entropy, the power spectrum entropy and the high-order cumulative quantity ratio characteristic parameters, and determining whether the signal is disturbed; the system, device and medium can identify the disturbed state of the PSK digital modulation communication signal; the identification is more comprehensive and accurate, and the accuracy of subsequent detection and positioning can be improved.

Description

Technical Field

[0001] This invention belongs to the field of communication signal anti-interference technology, specifically relating to a method, system, device and medium for identifying the disturbed state of PSK digital modulation communication signals. Background Technology

[0002] With the continuous development of communication technology, the electromagnetic environment is becoming increasingly complex, posing a threat to communication quality in wireless communication. Due to the openness of the electromagnetic space during communication, interference signals are inevitably received when receiving one's own communication signals, including various unintentional and malicious interferences. Unintentional interference mainly includes interference from natural phenomena and mutual interference between devices. Malicious interference refers to man-made radiated signals that disrupt, damage, and deceive the receiver, preventing it from obtaining effective information. Communication systems cannot avoid unintentional interference, but they can suppress it through certain means. However, malicious interference can seriously affect the communication system, leading to a deterioration in the quality of the received communication signals, making it impossible to obtain effective information, affecting the accuracy of subsequent detection, positioning, and other tasks, and thus failing to achieve the expected military objectives. Therefore, in-depth research on communication anti-interference technology is of great significance to wireless communication technology. Interference detection is a crucial link in anti-interference, its purpose being to determine whether interference exists in the received signal and to feed the results back to the transmitter or command center, which is essential for taking effective anti-interference measures.

[0003] The basic principles of several common methods for judging interference in existing communication signals are as follows: Energy detection method: First, two assumptions are made based on whether there is an interference signal. When there is no interference signal, the signal energy is calculated and a decision threshold is set. Then, the energy of the received signal is calculated and compared with the decision threshold value. If it is greater than the threshold value, the interference is judged to exist. Cyclic stationary detection method: The cyclostationary characteristics and correlation of the signal are used to judge whether there is interference.

[0004] Most existing interference detection algorithms are designed for situations where only interference signals and noise exist. A few technologies consider the communication signal itself or only consider a single communication signal. In practice, the sender will transmit a communication signal with a modulation type suitable for transmission based on the channel quality to achieve the purpose of transmitting effective information. If affected by enemy interference signals, the sender cannot obtain effective information, affecting the accuracy of subsequent detection, positioning, and other tasks. In addition, some technologies rely on the prior analysis of several interference signal characteristics to achieve the purpose of interference judgment, but the identification is not accurate enough, affecting the subsequent anti-interference effect.

[0005] Chinese patent CN112838909B proposes a communication interference detection method based on Gaussian eye diagram texture entropy features, mainly addressing the problems of low detection rate and long detection time in existing interference detection methods under low interference-to-signal ratio. The implementation steps are: (1) generating a Gaussian eye diagram of the wireless communication signal to be detected; (2) calculating the texture entropy features of the Gaussian eye diagram; (3) setting the interference detection test statistic; (4) performing the test decision to obtain the interference detection result. This invention has the advantages of high interference detection probability and fast detection speed under both high and low interference-to-signal ratios, effectively overcoming the problems of low detection probability and long detection time in existing interference detection methods under low interference-to-signal ratios. However, firstly, this patent only considers the presence of useful signals and interference signals; secondly, this patent uses the entropy of the eye diagram to identify the disturbed state of PSK-type signals, and cannot identify the disturbed states of BPSK, QPSK, and 8PSK signals. Summary of the Invention

[0006] To overcome the shortcomings of the existing technology, the present invention aims to propose a method, system, device, and medium for identifying the disturbed state of PSK digital modulation communication signals. By analyzing the characteristics of phase shift keying (PSK) communication signals, including the spectral symmetry, number of discrete peaks in the spectrum, time-domain amplitude entropy, power spectral entropy, and ratio of higher-order cumulants, the stable characteristic parameters of different modulation types and their stable ranges are determined to determine whether the signal is disturbed. This allows for rapid identification of disturbed information, more comprehensive and accurate identification, and improved accuracy of subsequent detection and positioning.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0008] A method for identifying the disturbed state of a PSK digital modulation communication signal, specifically including the following steps:

[0009] Step 1: Analyze the characteristic parameters of signals with different modulation types, including spectral symmetry P, the number of discrete peaks in the spectrum Peak-num, and the temporal amplitude entropy H. a Power spectral entropy H f The analysis of the characteristic parameters of the higher-order cumulant ratios F1, F2, and F8 yields characteristic parameters for identifying the disturbed state.

[0010] Step 2: Based on the characteristic parameters of spectral symmetry, number of discrete peaks in the spectrum, time-domain amplitude entropy, power spectral entropy, and ratio of higher-order cumulants, determine the stable range of PSK characteristic parameters when there is no interference, and determine whether the signal is disturbed.

[0011] The characteristic parameters of spectral symmetry in step one are analyzed as follows:

[0012] Spectral symmetry (P) refers to whether the spectrum of a communication signal is symmetrical about the carrier frequency.

[0013]

[0014] In the formula,

[0015]

[0016]

[0017] Among them, P U P L These represent the spectrum of the upper and lower sidebands of the signal, respectively, with the upper sideband divided by the carrier frequency;

[0018] When the signal spectrum is symmetrical, the theoretical value of the spectral symmetry P is 0. According to the fact that PSK signals have a symmetrical spectrum centered on the carrier frequency and maintain spectral symmetry when unaffected by interference, it can be concluded that when the signal is subjected to suppressive interference and the frequency of the interference is not at the signal carrier frequency, the spectral characteristics of the PSK signal spectrum change and the spectral symmetry is destroyed.

[0019] The analysis of the number of discrete peaks in the frequency domain, Peak-num, in step one is as follows:

[0020] First, the spectrum data is preprocessed: using {R(f i For each frequency point f of the group i = 1, 2, ..., N, ... i Spectral value at f i-A ~f i-B and f i+A ~f i+B The spectrum R'(f) is obtained by taking the ratio of the mean of the two m maximum values:

[0021] Then, spectral line extraction is performed to obtain the number of discrete peaks in the spectrum, Peak-num. When the spectrum of a PSK signal has no discrete spectrum and is not affected by interference, the number of spectral peaks in the PSK signal, Peak-num, is equal to 0.

[0022] In step one, the temporal magnitude entropy H is... a The analysis is as follows:

[0023] Let the received signal be r(t), and the sampling rate be f. s We obtain r(n) by sampling it, where n = 1, 2, ..., N, and N is the total number of sampling points; the instantaneous amplitude of the signal can be expressed as: a(n) = |r(n)|; the time-domain amplitude entropy of the signal can be obtained using the instantaneous amplitude of the signal.

[0024] Normalize the instantaneous amplitude a(n)

[0025]

[0026] The time-domain amplitude entropy of the signal is

[0027]

[0028] In the formula, This represents the proportion of each amplitude energy to the total energy.

[0029] Time-domain amplitude entropy H a The amplitude entropy characterizes the degree of dispersion of the amplitude energy of a digital signal. The more dispersed the amplitude energy, the greater the time-domain amplitude entropy; the more concentrated the amplitude energy, the smaller the value. PSK-type signals transmit information according to phase changes during modulation. Their amplitude energy is evenly distributed within the sampling interval, and the value of the amplitude entropy is stable when unaffected by interference.

[0030] In step one, the power spectral entropy H f The analysis is as follows:

[0031] The received signal, after discrete sampling, is r(n). The power spectrum X(w) of the signal is calculated using the periodogram method as follows:

[0032]

[0033] In the formula, R(w) is the Fourier transform of r(n), then the power spectral entropy of r(n) is:

[0034]

[0035] Power spectral entropy H f The power spectral entropy characterizes the degree of energy dispersion in a digital signal, reflecting the number of power spectral lines. A higher power spectral entropy indicates a greater number of spectral lines, while a lower entropy indicates fewer lines. For PSK-type signals, whose power spectrum only has continuous spectral components, their energy is more dispersed. When unaffected by interference, the H of PSK... f It is relatively stable, but if the signal is interfered with, its power spectrum energy dispersion will be disrupted, i.e., H f Fluctuations occurred.

[0036] The analysis of the higher-order cumulative ratios F1, F2, and F8 in step one is as follows:

[0037] The higher-order cumulants of a Gaussian random variable are zero, while the higher-order cumulants of a non-Gaussian random variable are not zero. Theoretically, the higher-order cumulants of a digital signal containing Gaussian white noise and a digital signal without noise are the same.

[0038] For a complex random process X(t) with zero mean, its higher-order moments are defined as:

[0039] M pq =E[X(t)] (p-q) X* (t) q (1-8)

[0040] Cumulative amount is defined as:

[0041] C pq =C um {X(t),···,X(t),X * (t),···,X * (t)} (1-9)

[0042] Where X(t) is the pq term, X * (t) is term q, C um For cumulative moments, * denotes conjugate;

[0043] The relationship between cumulants of each order and moments is as follows:

[0044] C 20 =M 20 (1-10)

[0045] C 21 =M 21 (1-11)

[0046] C 40 =M 40 -3M 20 2 (1-12)

[0047] C 42 =M 42 -|M 20 | 2 -2M 21 2 (1-13)

[0048] C 60 =M 60 -15M 40 M 20 +30M 20 3 (1-14)

[0049]

[0050] Since the signal s(t) and the Gaussian white noise n(t) are independent, according to the properties of cumulants, we can obtain:

[0051] C um (r(t))=C um (s(t))+C um (n(t)) (1-16)

[0052] Since the cumulative value of zero-mean Gaussian white noise greater than second-order is zero, the above formula can be expressed as: C um (r(t))=C um (s(t));

[0053] When a modulated signal passes through a Gaussian channel, due to the superposition of noise, the theoretical model for the received signal is as follows:

[0054]

[0055] Where E is the modulation energy of the symbol, and a k T is the energy-normalized digital symbol sequence. s For the symbol period, ω c Let θ be the carrier frequency, θ be the initial phase of the carrier, and the symbol values ​​occur with equal probability; n(t) is zero-mean Gaussian white noise; after the phase, carrier frequency, and timing of the received signal are synchronized, the received signal is down-converted and sampled at the correct time. The theoretical higher-order cumulative quantities of the PSK type signal are shown in Table 1.

[0056] Table 1 Theoretical higher-order cumulants for PSK-type signals

[0057] |C20| |C21| |C40| |C41| |C42| |C60| |C63| BPSK E E <![CDATA[2E 2 ]]> <![CDATA[2E 2 ]]> <![CDATA[2E 2 ]]> <![CDATA[16E 3 ]]> <![CDATA[13E 3 ]]> QPSK 0 E <![CDATA[E 2 ]]> 0 <![CDATA[E 2 ]]> 0 <![CDATA[4E 3 ]]> 8PSK 0 0 0 0 <![CDATA[E 2 ]]> 0 <![CDATA[4E 3 ]]>

[0058] In Table 1, all higher-order cumulants are related to E, and signals with the same modulation scheme exhibit differences in their higher-order cumulants. Characteristic parameters related to the ratio of higher-order cumulants are constructed to eliminate the influence of E on the ratio of higher-order cumulants among signals of the same modulation type.

[0059] in,

[0060] Step two, based on the characteristic parameters of spectral symmetry, the number of discrete peaks in the spectrum, time-domain amplitude entropy, power spectral entropy, and the ratio of higher-order cumulants, determines the stable range of PSK characteristic parameters when there is no interference, and determines whether the signal is disturbed, as follows:

[0061] 1) Determine the stability range of PSK-type signal characteristic parameters when there is no interference:

[0062] The stable range of characteristic parameters of BPSK, QPSK, and 8PSK signals under interference-free conditions was calculated and determined when the bit signal-to-noise ratio (SNR) was 8-20 dB. At least 50 characteristic parameter calculations were performed at each SNR, and the average value was taken as the characteristic parameter value at that SNR. Symbols were randomly generated when generating BPSK, QPSK, and 8PSK baseband signals, and the oversampling rate of the baseband signal was 8. The spectral symmetry P, the number of discrete peaks in the spectrum Peak-num, and the temporal amplitude entropy H were obtained. a Power spectral entropy H fThe stability range of the characteristic parameters of the higher-order cumulant ratios F1, F2, and F8 within a bit signal-to-noise ratio of 8-20 dB is shown in Table 2.

[0063] Table 2 shows the stability range of PSK signal characteristic parameters under interference-free conditions.

[0064] Peak_num P <![CDATA[H a ]]> <![CDATA[H f ]]> <![CDATA[F1]]> <![CDATA[F2]]> <![CDATA[F8]]> BPSK 0 (-0.05,0.05) (4.4,4.55) (3.6,4.5) (0.9,1.1) - - QPSK 0 (-0.05,0.05) (4.5,4.6) (3.6,4.5) (0.9,1.2) (0.6,1.05) - 8PSK 0 (-0.05,0.05) (4.5,4.6) (3.0,4.4) - - (14.6,15.5)

[0065] 2) Using the stable range of PSK characteristic parameters in Table 2 when there is no interference as the basis for judging whether the signal is disturbed, spectral symmetry, number of discrete peaks in the spectrum, time-domain amplitude entropy, and power spectrum entropy are selected as characteristic parameters for judging disturbance for PSK type signals. For BPSK modulation type, in addition to spectral symmetry, number of discrete peaks in the spectrum, time-domain amplitude entropy, and power spectrum entropy, F1 is selected as a characteristic parameter for judging disturbance. For QPSK, in addition to spectral symmetry, number of discrete peaks in the spectrum, time-domain amplitude entropy, and power spectrum entropy, F1 and F2 are selected as characteristic parameters for judging disturbance. For 8PSK, in addition to spectral symmetry, number of discrete peaks in the spectrum, time-domain amplitude entropy, and power spectrum entropy, F1 and F8 are selected as characteristic parameters unique to its disturbance identification. When one or more characteristic parameter values ​​exceed the range, the signal is judged to be disturbed.

[0066] The disturbance state identification system for PSK digital modulation communication signals based on the above identification method includes:

[0067] Input module: Used to provide the prior conditions required by the system, i.e. the signal modulation method; input the sampled data of the received signal.

[0068] Feature parameter calculation module: Calculates the feature parameters of the received signal for subsequent disturbance state identification.

[0069] Output module: Utilizes the modulation method of the input signal, calls the corresponding disturbance state recognition algorithm, and outputs the disturbance state of the signal.

[0070] The disturbance status identification device for PSK digital modulation communication signals based on the above identification method includes:

[0071] Memory, used to store computer programs;

[0072] A processor is used to implement the method for identifying the disturbed state of PSK digital modulation communication signals as described in any one of steps one or two when executing the computer program.

[0073] A computer-readable storage medium storing a computer program that, when executed by a processor, can identify the disturbance state of a PSK digital modulation communication signal.

[0074] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0075] 1. The method proposed in this invention takes into account the disturbance state identification when communication signals, noise and interference signals coexist. When the user only knows that the interference signal belongs to the PSK type, a disturbance state identification method that is universal for PSK type signals is designed. At the same time, disturbance state identification can be performed separately for BPSK, QPSK and 8PSK signals. It can quickly obtain disturbance information, has strong practicality, and improves the accuracy of subsequent detection and positioning.

[0076] 2. P, Peak-num, H a and H f These are the characteristic parameters that reflect the stability of PSK-type signals. When a PSK-type signal is interfered with, the values ​​of at least one or more of these characteristic parameters will change. When a PSK-type signal is subjected to single-tone or multi-tone interference, the number of spectral peaks in the signal will change. When a PSK-type signal is subjected to pulse interference, due to the characteristics of pulse interference, the signal will exhibit burst characteristics in the time domain and spectral broadening in the frequency domain, thus affecting the P and H values ​​of the PSK-type signal. a and H f Significant changes; when PSK-type signals are subjected to linear frequency modulation interference and narrowband interference, the signal frequency domain is affected, impacting the signal's H... f It has an impact; when subjected to signal interference, the interference signal causes changes in the higher-order cumulative information of the signal, and the characteristic parameters F1, F2 and F8 change compared to when they are not disturbed; the identification is more comprehensive and accurate.

[0077] 3. Through extensive simulation experiments, the interference detection algorithm of this invention achieves an accuracy rate of over 98% for each modulation type when the bit signal-to-noise ratio is above 5dB and the interference-to-signal ratio is above -3dB; for the interference detection algorithm of PSK modulation type, the accuracy rate of interference detection is over 98% when the bit signal-to-noise ratio is above 5dB and the interference-to-signal ratio is above -1dB. Attached Figure Description

[0078] Figure 1 These are images showing the effects before and after spectrum preprocessing, where: Figure 1 (a) is the effect before spectrum preprocessing. Figure 1 (b) is the effect diagram before spectrum preprocessing.

[0079] Figure 2 These are curves showing the variation of characteristic parameters of BPSK, QPSK, and 8PSK with signal-to-noise ratio, where: Figure 2 (a) shows the curves of BPSK characteristic parameters as a function of signal-to-noise ratio. Figure 2 (b) shows the curves of QPSK characteristic parameters as a function of signal-to-noise ratio. Figure 2 (c) shows the curves of 8PSK characteristic parameters as a function of signal-to-noise ratio.

[0080] Figure 3 This is a time-domain and frequency-domain waveform diagram of a BPSK after being subjected to single-tone interference, where: Figure 3 (a) is the time-domain waveform of BPSK after being subjected to single-tone interference. Figure 3 (b) is the frequency domain waveform of BPSK after being subjected to single-tone interference. Figure 3 (c) is the time-domain waveform of BPSK after being subjected to multi-tone interference. Figure 3 (d) is the frequency domain waveform of BPSK after being subjected to multi-tone interference. Figure 3 (e) is the time-domain waveform of BPSK after being subjected to linear frequency modulation interference. Figure 3 (f) is the frequency domain waveform of BPSK after being subjected to linear frequency modulation interference. Figure 3 (g) is the time-domain waveform of BPSK after narrowband interference. Figure 3 (h) is the frequency domain waveform of BPSK after narrowband interference. Figure 3 (i) is the time-domain waveform of BPSK after being subjected to pulse interference. Figure 3 (j) is the frequency domain waveform of BPSK after being subjected to pulse interference.

[0081] Figure 4 This is the parameter variation curve of the BPSK signal under single-tone interference, where: Figure 4 (a) is the parameter H of the BPSK signal affected by single-tone interference. a Change curve, Figure 4 (b) is the parameter H of the BPSK signal affected by single-tone interference. f Change curve, Figure 4 (c) shows the curve of parameter P changing due to single-tone interference on the BPSK signal. Figure 4 (d) shows the Peak-num variation curve of the BPSK signal under single-tone interference. Figure 4 (e) is the curve of the variation of the higher-order cumulative characteristic F1 of the BPSK signal under single-tone interference.

[0082] Figure 5 This is the parameter variation curve of the BPSK signal under multi-tone interference, where: Figure 5 (a) is the parameter H of the BPSK signal affected by single-tone interference. a Change curve, Figure 5 (b) is the parameter H of the BPSK signal affected by single-tone interference. f Change curve, Figure 5 (c) shows the curve of parameter P changing due to single-tone interference on the BPSK signal. Figure 5 (d) shows the Peak-num variation curve of the BPSK signal under single-tone interference. Figure 5 (e) is the curve of the variation of the higher-order cumulative characteristic F1 of the BPSK signal under single-tone interference.

[0083] Figure 6 This is the parameter variation curve of the BPSK signal under linear frequency modulation interference, where: Figure 6 (a) is the parameter H of the BPSK signal subjected to linear frequency modulation interference. a Change curve, Figure 6 (b) is the parameter H of the BPSK signal subjected to linear frequency modulation interference. f Change curve, Figure 6 (c) shows the curve of parameter P changing under linear frequency modulation interference on the BPSK signal. Figure 6 (d) shows the Peak_num variation curve of the BPSK signal under linear frequency modulation interference. Figure 6 (e) is the curve of the variation of the higher-order cumulative characteristic F1 of the BPSK signal under single-tone interference.

[0084] Figure 7 This is the parameter variation curve of the BPSK signal under narrowband interference, where: Figure 7 (a) is the parameter H of the BPSK signal affected by narrowband interference. a Change curve, Figure 7 (b) is the parameter H of the BPSK signal affected by narrowband interference. f Change curve, Figure 7 (c) shows the curve of parameter P changing due to narrowband interference in the BPSK signal. Figure 7 (d) shows the Peak_num variation curve of the BPSK signal under narrowband interference. Figure 7 (e) is the curve of the variation of F1, a high-order cumulative characteristic of the BPSK signal under narrowband interference.

[0085] Figure 8 This is the parameter variation curve of the BPSK signal under pulse interference, where: Figure 8 (a) is the pulse interference parameter H of the BPSK signal. a Change curve, Figure 8 (b) is the pulse interference parameter H of the BPSK signal. f Change curve, Figure 8 (c) shows the curve of parameter P changing under pulse interference on the BPSK signal. Figure 8 (d) shows the peak_num variation curve of the BPSK signal under pulse interference. Figure 8 (e) is the curve of the change of F1, a higher-order cumulative characteristic of the BPSK signal under pulse interference.

[0086] Figure 9 This is the parameter variation curve of a QPSK signal affected by single-tone interference, where: Figure 9 (a) is the parameter H of the QPSK signal affected by single-tone interference. a Change curve, Figure 9 (b) is the parameter H of the QPSK signal affected by single-tone interference. f Change curve, Figure 9 (c) shows the curve of the QPSK signal affected by single-tone interference parameter F1. Figure 9(d) is the curve of F2 variation of QPSK signal under single-tone interference.

[0087] Figure 10 This is the parameter variation curve of the QPSK signal under multi-tone interference, where: Figure 10 (a) is the parameter H of the QPSK signal affected by multitone interference. a Change curve, Figure 10 (b) is the parameter H of the QPSK signal affected by multitone interference. f Change curve, Figure 10 (c) shows the variation curve of parameter F1 of QPSK signal under multi-tone interference. Figure 10 (d) is the curve of F2 variation of QPSK signal under multi-tone interference.

[0088] Figure 11 This is the parameter variation curve of the QPSK signal under linear frequency modulation interference, where: Figure 11 (a) is the parameter H of the QPSK signal subjected to linear frequency modulation interference. a Change curve, Figure 11 (b) is the parameter H of the QPSK signal subjected to linear frequency modulation interference. f Change curve, Figure 11 (c) shows the variation curve of parameter F1 of QPSK signal under linear frequency modulation interference. Figure 11 (d) is the curve of F2 variation of QPSK signal under linear frequency modulation interference.

[0089] Figure 12 This is the parameter variation curve of the QPSK signal under narrowband interference, where: Figure 12 (a) is the parameter H of the QPSK signal affected by narrowband interference. a Change curve, Figure 12 (b) is the parameter H of the QPSK signal affected by narrowband interference. f Change curve, Figure 12 (c) shows the variation curve of parameter F1 of QPSK signal under narrowband interference. Figure 12 (d) is the curve of F2 variation of QPSK signal under narrowband interference.

[0090] Figure 13 This is the parameter variation curve of the QPSK signal under pulse interference, where: Figure 13 (a) is the pulse interference parameter H of the QPSK signal. a Change curve, Figure 13 (b) is the pulse interference parameter H of the QPSK signal. f Change curve, Figure 13 (c) shows the curve of the QPSK signal under pulse interference parameter F1. Figure 13 (d) is the curve of the QPSK signal changing under pulse interference F2.

[0091] Figure 14This is the curve showing the change of characteristic parameters of an 8PSK signal under single-tone interference, where: Figure 14 (a) is the parameter H of the 8PSK signal affected by single-tone interference. a Change curve, Figure 14 (b) is the parameter H of the 8PSK signal affected by single-tone interference. f Change curve, Figure 14 (c) shows the curve of the parameter F8 changing due to single-tone interference of the 8PSK signal.

[0092] Figure 15 This is the characteristic parameter variation curve of an 8PSK signal subjected to multi-tone interference, where: Figure 15 (a) is the parameter H of the 8PSK signal affected by multitone interference. a Change curve, Figure 15 (b) is the parameter H of the 8PSK signal affected by multitone interference. f Change curve, Figure 15 (c) shows the curve of the variation of parameter F8 of the 8PSK signal under multi-tone interference.

[0093] Figure 16 This is the curve showing the change of characteristic parameters of an 8PSK signal subjected to linear frequency modulation interference, where: Figure 16 (a) is the parameter H of the 8PSK signal subjected to linear frequency modulation interference. a Change curve, Figure 16 (b) is the 8PSK signal subjected to linear frequency modulation parameter H f Change curve, Figure 16 (c) is the curve of the variation of parameter F8 of the 8PSK signal under linear frequency modulation interference.

[0094] Figure 17 This is the curve showing the variation of characteristic parameters of an 8PSK signal under narrowband interference, where: Figure 17 (a) is the parameter H of the 8PSK signal subjected to narrowband interference. a Change curve, Figure 17 (b) is the parameter H of the 8PSK signal subjected to narrowband interference. f Change curve, Figure 17 (c) shows the variation curve of parameter F8 of the 8PSK signal under narrowband interference.

[0095] Figure 18 This is the curve showing the change of characteristic parameters of an 8PSK signal under pulse interference, where: Figure 18 (a) is the pulse interference parameter H of the 8PSK signal. a Change curve, Figure 18 (b) is the pulse interference parameter H of the 8PSK signal. f Change curve, Figure 18 (c) shows the curve of parameter F8 changing under pulse interference on the 8PSK signal.

[0096] Figure 19This refers to the accuracy of PSK-based disturbed state recognition algorithms, where: Figure 19 (a) Based on decision trees, Figure 19 (b) Based on SVM.

[0097] Figure 20 The accuracy of the algorithm for identifying the disturbed state of different modulated signals based on decision trees is as follows: Figure 20 (a) represents the accuracy of the BPSK signal disturbance state identification algorithm. Figure 20 (b) represents the accuracy of the QPSK signal disturbance state identification algorithm. Figure 20 (c) represents the accuracy of the QPSK signal disturbance state identification algorithm.

[0098] Figure 21 The accuracy of the SVM-based algorithm for identifying the disturbed states of different modulated signals is as follows: Figure 21 (a) represents the accuracy of the BPSK signal disturbance state identification algorithm. Figure 21 (b) represents the accuracy of the QPSK signal disturbance state identification algorithm. Figure 21 (c) represents the accuracy of the QPSK signal disturbance state identification algorithm. Detailed Implementation

[0099] The present invention will now be described in further detail with reference to the accompanying drawings.

[0100] A method for identifying the disturbed state of a PSK digital modulation communication signal, specifically including the following steps:

[0101] Step 1: Analyze the characteristic parameters of signals with different modulation types, and select the characteristic parameters for disturbance state identification. The characteristic parameters and analysis are as follows:

[0102] (1) Spectral symmetry P-analysis

[0103] Spectral symmetry refers to whether the spectrum of a communication signal is symmetrical about the carrier frequency.

[0104]

[0105] In the formula,

[0106]

[0107]

[0108] Among them, P U P L These represent the spectrum of the upper and lower sidebands of the signal, respectively, with the upper sideband divided by the carrier frequency;

[0109] When the signal spectrum is symmetrical, the theoretical value of the spectral symmetry P is 0. According to the fact that PSK signals have a symmetrical spectrum centered on the carrier frequency, the spectral symmetry is maintained when there is no interference. When the signal is subjected to suppressive interference and the frequency of the interference is not at the signal carrier frequency, the spectral characteristics of the PSK signal spectrum change and the spectral symmetry is destroyed.

[0110] (2) Peak-num analysis, which is the number of discrete peaks in the frequency domain of the signal;

[0111] First, the spectrum data is preprocessed: using {R(f i For each frequency point f of the group i = 1, 2, ..., N, ... i Spectral value at f i-A ~f i-B and f i+A ~f i+B The spectrum R'(f) is obtained by taking the ratio of the mean of the two m maximum values.

[0112] Then, spectral line extraction is performed to obtain the number of discrete peaks in the spectrum, Peak-num. PSK signals have no discrete spectrum; when unaffected by interference, the number of peaks in the PSK signal, Peak-num, is equal to 0. The resulting graph is shown below. Figure 1 ;

[0113] (3) Temporal amplitude entropy H a analyze;

[0114] Let the received signal be r(t), and the sampling rate be f. s We obtain r(n) by sampling it, where n = 1, 2, ..., N, and N is the total number of sampling points; the instantaneous amplitude of the signal can be expressed as: a(n) = |r(n)|; the time-domain amplitude entropy of the signal can be obtained using the instantaneous amplitude of the signal.

[0115] Normalize the instantaneous amplitude a(n)

[0116]

[0117] The time-domain amplitude entropy of the signal is

[0118]

[0119] In the formula This represents the proportion of each amplitude energy to the total energy.

[0120] Time-domain amplitude entropy H aThe amplitude entropy characterizes the degree of dispersion of the amplitude energy of a digital signal. The more dispersed the amplitude energy, the greater the time domain amplitude entropy; the more concentrated the amplitude energy, the smaller the value. PSK-type signals transmit information according to phase changes during modulation. Their amplitude energy is evenly distributed within the sampling interval, and the value of amplitude entropy is stable when unaffected by interference.

[0121] (4) Power spectral entropy H f analyze

[0122] The received signal, after discrete sampling, is r(n). The power spectrum X(w) of the signal is calculated using the periodogram method as follows:

[0123]

[0124] In the formula, R(w) is the Fourier transform of r(n), then the power spectral entropy of r(n) is:

[0125]

[0126] Power spectral entropy H f The power spectral entropy characterizes the degree of energy dispersion in a digital signal, reflecting the number of power spectral lines. A higher power spectral entropy indicates a greater number of spectral lines, while a lower entropy indicates fewer lines. For PSK-type signals, whose power spectrum only has continuous spectral components, their energy is relatively dispersed. When unaffected by interference, the H of PSK... f It is relatively stable, but if the signal is interfered with, its power spectrum energy dispersion will be disrupted, i.e., H f The signal will fluctuate, and if the fluctuation exceeds the stable range of the signal, it proves that there is interference.

[0127] (5) Analysis of higher-order cumulative ratios F1, F2, and F8

[0128] The higher-order cumulants of Gaussian random variables are zero, while the higher-order cumulants of non-Gaussian random variables are not zero. Theoretically, the higher-order cumulants of digital signals containing Gaussian white noise and those without noise are the same. Extracting the higher-order cumulants of noisy signals as signal features can minimize the impact of noise on the signal.

[0129] For a complex random process X(t) with zero mean, its higher-order moments are defined as:

[0130] M pq =E[X(t)] (p-q) X * (t) q (1-8)

[0131] Cumulative amount is defined as:

[0132] C pq =C um{X(t),···,X(t),X * (t),···,X * (t)} (1-9)

[0133] Where X(t) is the pq term, X * (t) is term q, C um For cumulative moments, * denotes conjugate;

[0134] The relationship between cumulants of each order and moments is as follows:

[0135] C 20 =M 20 (1-10)

[0136] C 21 =M 21 (1-11)

[0137] C 40 =M 40 -3M 20 2 (1-12)

[0138] C 42 =M 42 -|M 20 | 2 -2M 21 2 (1-13)

[0139] C 60 =M 60 -15M 40 M 20 +30M 20 3 (1-14)

[0140]

[0141] Since the signal s(t) and the Gaussian white noise n(t) are independent, according to the properties of cumulants, we can obtain:

[0142] C um (r(t))=C um (s(t))+C um (n(t)) (1-16)

[0143] Since the cumulative value of zero-mean Gaussian white noise greater than second order is zero, the above formula can be expressed as: C um (r(t))=C um (s(t));

[0144] When a modulated signal passes through a Gaussian channel, due to the superposition of noise, the theoretical model for the received signal is as follows:

[0145]

[0146] Where E is the modulation energy of the symbol, and a k T is the energy-normalized digital symbol sequence. s For the symbol period, ω c Let θ be the carrier frequency, θ be the initial phase of the carrier, where the symbol values ​​occur with equal probability; n(t) is zero-mean Gaussian white noise; after the phase, carrier frequency, and timing of the received signal are synchronized, the received signal is down-converted and sampled at the correct time. The theoretical higher-order cumulants of the PSK class signal are as follows:

[0147] Table 1 Theoretical higher-order cumulants for PSK-type signals

[0148] |C20| |C21| |C40| |C41| |C42| |C60| |C63| BPSK E E <![CDATA[2E 2 ]]> <![CDATA[2E 2 ]]> <![CDATA[2E 2 ]]> <![CDATA[16E 3 ]]> <![CDATA[13E 3 ]]> QPSK 0 E <![CDATA[E 2 ]]> 0 <![CDATA[E 2 ]]> 0 <![CDATA[4E 3 ]]> 8PSK 0 0 0 0 <![CDATA[E 2 ]]> 0 <![CDATA[4E 3 ]]>

[0149] In Table 1, all higher-order cumulants are related to E, and signals with the same modulation scheme exhibit differences in their higher-order cumulants. Characteristic parameters related to the ratio of higher-order cumulants are constructed to eliminate the influence of E on the ratio of higher-order cumulants among signals of the same modulation type.

[0150] in,

[0151] 2. Based on the characteristic parameters of spectral symmetry, number of discrete peaks in the spectrum, time-domain amplitude entropy, power spectral entropy, and ratio of higher-order cumulants, determine the stable range of PSK characteristic parameters when there is no interference, and determine whether the signal is disturbed.

[0152] 1) Determine the stable range of PSK characteristic parameters when there is no disturbance:

[0153] The stable range of characteristic parameters of BPSK, QPSK, and 8PSK signals under interference-free conditions was calculated and determined at bit signal-to-noise ratios (SNRs) of 8-20 dB in 2 dB increments. At least 50 characteristic parameter calculations were performed at each SNR, and the average value was taken as the characteristic parameter value at that SNR. Symbols were randomly generated when generating BPSK, QPSK, and 8PSK baseband signals. The digital communication signal was a baseband signal with a code rate of 0.5 Mbps, an adjacent frequency interval of twice the code rate, and a root-raised cosine filter with a roll-off factor of 0.5. The oversampling rate of the baseband signal was 8. The spectral symmetry P, the number of discrete peaks Peak-num, and the temporal amplitude entropy H were obtained. a Power spectral entropy H f The average of the higher-order cumulant ratios F1, F2, and F8 is taken as the stable range of characteristic parameters for this type of signal under this SNR (signal-to-noise ratio) condition:

[0154] Table 2 shows the stability range of PSK signal characteristic parameters under interference-free conditions.

[0155] Peak_num P <![CDATA[H a ]]> <![CDATA[H f ]]> <![CDATA[F1]]> <![CDATA[F2]]> <![CDATA[F8]]> BPSK 0 (-0.05,0.05) (4.4,4.55) (3.6,4.5) (0.9,1.1) - - QPSK 0 (-0.05,0.05) (4.5,4.6) (3.6,4.5) (0.9,1.2) (0.6,1.05) - 8PSK 0 (-0.05,0.05) (4.5,4.6) (3.0,4.4) - - (14.6,15.5)

[0156] 2) Using the stable range of PSK characteristic parameters in Table 2 when there is no interference as the basis for judging whether the signal is disturbed, spectral symmetry, number of discrete peaks in the spectrum, temporal amplitude entropy, and power spectral entropy are selected as characteristic parameters for disturbance judgment for all PSK signals. For BPSK modulation type, in addition to spectral symmetry, number of discrete peaks in the spectrum, temporal amplitude entropy, and power spectral entropy, F1 is selected as a characteristic parameter for disturbance judgment. For QPSK, in addition to spectral symmetry, number of discrete peaks in the spectrum, temporal amplitude entropy, and power spectral entropy, F1 and F2 are selected as characteristic parameters for disturbance judgment. For 8PSK, in addition to spectral symmetry, number of discrete peaks in the spectrum, temporal amplitude entropy, and power spectral entropy, F1 and F8 are selected as characteristic parameters specific to its disturbance identification. When one or more characteristic parameter values ​​exceed the range, the signal is judged to be disturbed. The curves of BPSK, QPSK, and 8PSK characteristic parameters changing with signal-to-noise ratio are shown in the figure. Figure 2 .

[0157] The following section analyzes the disturbance of characteristic parameters based on simulation experiments.

[0158] 1. Changes in the perturbation characteristic parameters of the BPSK signal

[0159] analyze Figure 2 The disturbance changes of stable characteristic parameters are simulated and analyzed using single-tone interference, multi-tone interference, linear frequency modulation interference, narrowband interference, and pulse interference.

[0160] (1) The effect of single-tone interference on the characteristic parameters of BPSK

[0161] The BPSK signal is affected by a single-tone interference (STO), with an STO-to-MS ratio of 8 dB and a STO frequency of 2 kHz. The time-domain and frequency-domain waveforms of the signal are shown below. Figure 3 :

[0162] Simulate the changes in the characteristic parameters of the BPSK signal under single-tone interference, such as... Figure 4 .

[0163] The frequency domain characteristic of single-tone interference is that it has a discrete spectrum at the signal frequency. Since the spectrum of a BPSK signal does not contain a discrete spectrum, a discrete spectral line will appear in the spectrum of a BPSK signal after being subjected to single-tone interference, making Peak-num = 1. Simultaneously, this affects the BPSK signal H... f This has an impact; as the signal-to-interference ratio increases, H... f Reduce. Single-tone interference affects the H of the signal. a Yes, it has an impact. Single-tone interference disperses the time-domain energy of the signal, and this dispersion becomes even more pronounced as the interference power increases. aThe signal intensity increases, but its fluctuations are small. Single-tone interference significantly affects the higher-order cumulant characteristic F1 of the BPSK signal. Single-tone interference has almost no effect on the signal spectral symmetry P.

[0164] (2) The influence of multitone interference on the characteristic parameters of BPSK

[0165] The BPSK signal is affected by multi-tone interference, with an interference-to-signal ratio of 8 dB. The multi-tone interference frequencies are 10 kHz, 100 kHz, and 150 kHz. The time-domain and frequency-domain waveforms of the signal are shown below. Figure 3 :

[0166] Simulate the changes in four selected characteristic parameters of a BPSK signal (signal bit rate 0.5 Mbps, sampling rate 4 MHz, signal duration 10 ms, carrier frequency 1 kHz) under multi-tone interference. The frequencies of the multi-tone interference signals are 10 kHz, 100 kHz, and 150 kHz.

[0167] Multitone interference affects the time-domain amplitude of a signal, thus changing its time-domain energy entropy. Multitone interference concentrates the signal's time-domain energy, reducing the time-domain amplitude entropy, but the change is relatively small. Because multitone interference has multiple discrete spectra at the signal frequency, while the BPSK spectrum does not contain discrete spectra, multiple discrete spectral lines appear in BPSK after being subjected to multitone interference. Multitone interference significantly affects the higher-order cumulant characteristic F1 of BPSK. Multitone interference has almost no effect on signal symmetry. Figure 5 .

[0168] (3) The influence of linear frequency modulated signal on the characteristic parameters of BPSK

[0169] The BPSK signal is interfered with by a linear frequency modulated (LFM) signal, with an interference-to-signal ratio (ISR) of 8 dB between the interference and the BPSK signal. The time-domain and frequency-domain waveforms of the signal are shown below. Figure 3 .

[0170] Simulate the changes in four selected characteristic parameters of a BPSK signal (signal code rate 0.5 Mbps, sampling rate 4 MHz, signal duration 10 ms, carrier frequency 1 kHz) under linear frequency sweep interference. The frequency sweep rate of the linear frequency modulated signal is 100 MHz / s, and the starting frequency is 1 kHz.

[0171] The waveform of a linear frequency modulated (LFM) signal is approximately rectangular in the frequency domain and does not contain a discrete spectrum; therefore, the signal spectrum after interference does not have a discrete spectrum. LFM exhibits a wide bandwidth interference pattern, thus significantly affecting the spectral symmetry of the signal and its power spectral entropy. LFM interference affects the time-domain energy entropy of the signal, with parameter values ​​changing less than when there is no interference. It has a significant impact on the higher-order cumulants of the signal. For example... Figure 6 .

[0172] (4) The influence of narrowband interference on the characteristic parameters of BPSK

[0173] The BPSK signal is interfered with by a linear frequency modulated (LFM) signal, with an interference-to-signal ratio (ISR) of 8 dB between the interference and the BPSK signal. The time-domain and frequency-domain waveforms of the signal are shown below. Figure 3 :

[0174] Simulate the changes in four selected characteristic parameters of a BPSK signal (signal bit rate 0.5 Mbps, sampling rate 4 MHz, signal duration 10 ms, carrier frequency 1 kHz) under narrowband interference. The narrowband signal is a BPSK modulated signal with a bit rate of 10 Kbps and a carrier frequency of 5 kHz.

[0175] Since this narrowband interference is also a BPSK modulated signal, it has little impact on the time-domain amplitude entropy of the BPSK communication signal, nor does it affect the number of discrete spectral lines in the frequency domain. The interference signal has a very narrow bandwidth and almost no effect on the signal's spectral symmetry, but it has a significant impact on the power spectral entropy and higher-order cumulant F1 of the BPSK signal. Figure 7 .

[0176] (5) The effect of pulse interference on the characteristic parameters of BPSK

[0177] The BPSK signal is interfered with by a linear frequency modulated (LFM) signal, with an interference-to-signal ratio (ISR) of 8 dB between the interference and the BPSK signal. The time-domain and frequency-domain waveforms of the signal are shown below. Figure 3 :

[0178] Simulate the changes in four selected characteristic parameters of a BPSK signal (signal code rate 0.5 Mbps, sampling rate 4 MHz, signal duration 10 ms, carrier frequency 1 kHz) under pulse interference. The pulse interference is a triangular pulse with a pulse count of 4.

[0179] Because impulse interference is sudden and lasts for a period of time in the time domain, it has a significant impact on the time-domain amplitude entropy of the signal. Since the signal energy is concentrated in the time domain, it exhibits frequency broadening when transformed to the frequency domain, thus affecting the symmetry and power spectral entropy of the communication signal, but not causing the communication signal's spectrum to have a discrete spectrum. It also has a significant impact on the higher-order cumulant characteristic F1 of the signal. For example... Figure 8 As shown.

[0180] 2. Changes in the disturbance characteristic parameters of the QPSK signal; QPSK time-domain and frequency-domain waveforms are shown below. Figure 13 ;

[0181] Since QPSK and BPSK share the same spectral characteristics—both are continuous spectra without discrete spectral lines—and only single-tone and multi-tone interference affect the number of discrete peaks in the signal spectrum, the analysis of the impact of QPSK disturbances on characteristic parameters will no longer analyze the number of spectral peaks. Based on the above analysis of BPSK signal disturbances, it can be seen that spectral symmetry is affected by linear frequency modulation (LFM) interference and pulse interference. Therefore, the analysis of QPSK characteristic parameter values ​​affected by disturbances will only consider H... a H f F1, F2.

[0182] (1) The effect of single-tone interference on the characteristic parameters of QPSK

[0183] The QPSK signal is subject to single-tone interference, with an interference-to-signal ratio of 8 dB and a single-tone interference frequency of 2 kHz.

[0184] Simulate the changes of four selected characteristic parameters of a QPSK signal (carrier frequency 1kHz) under single-tone interference, with the frequency of the single-tone interference signal being 0.2kHz.

[0185] One-tone interference (MTI) affects the time-domain energy entropy of QPSK signals. As the interference power increases, the time-domain energy becomes more dispersed, and the time-domain amplitude entropy increases, but its fluctuation is relatively small. MTI significantly affects the higher-order cumulant characteristics (F1) of QPSK signals; for F2, the interference-to-signal ratio needs to be greater than 8 dB to be distinguishable from the signal. MTI has almost no effect on signal symmetry. Figure 9 .

[0186] (2) The effect of multitone interference on the characteristic parameters of QPSK

[0187] The QPSK signal is subject to multi-tone interference, with the interference having an interference-to-signal ratio of 8 dB.

[0188] Simulate the changes of four selected characteristic parameters of a QPSK signal (carrier frequency 1KHz) under multi-tone interference.

[0189] Multitone interference affects the time-domain energy entropy of the QPSK signal, but its fluctuation is relatively small. Multitone interference has a significant impact on the higher-order cumulant characteristics F1 and F2 of QPSK. Multitone interference affects the signal power spectral entropy, and H increases with the increase of the interference-to-signal ratio. f Decrease, such as Figure 10 As shown.

[0190] (3) The influence of linear frequency modulation signal on the characteristic parameters of QPSK

[0191] The QPSK signal is interfered with by a linear frequency modulated signal, with the interference-to-signal ratio of the QPSK signal being 8 dB.

[0192] Simulate the changes of four selected characteristic parameters of a QPSK signal (carrier frequency 1kHz) under linear frequency sweep interference. The frequency sweep rate of the linear frequency modulated signal is 100MHz / s, and the starting frequency is 1kHz.

[0193] Linear frequency modulation (LFM) interference affects the time-domain energy entropy of QPSK signals, with relatively small fluctuations in parameter values. However, it significantly impacts the signal power spectral entropy, and this effect increases with the increase of the interference-to-signal ratio (H). f Increased. Linear frequency modulation exhibits a wider bandwidth interference pattern, thus significantly affecting the spectral symmetry of the signal. It also has a significant impact on the higher-order cumulant parameter F1 of the signal. For example... Figure 11 .

[0194] (4) The impact of narrowband interference on QPSK characteristic parameters

[0195] The QPSK signal is subject to narrowband signal interference, with the interference-to-signal ratio of the QPSK signal being 8 dB.

[0196] The variations of four selected characteristic parameters of a QPSK signal (carrier frequency 1kHz) under narrowband interference were simulated. The narrowband signal was a QPSK modulated signal with a code rate of 10Kbps and a carrier frequency of 5kHz.

[0197] Narrowband interference has a relatively small impact on the time-domain amplitude entropy of QPSK communication signals. However, because narrowband signals result in a narrow spike in the signal power spectrum, they have a significant impact on the signal power spectral entropy, and this impact increases with the increase of the interference-to-signal ratio (H). f This is a significant increase. It has a substantial impact on the higher-order cumulant parameter F1 of the communication signal. For example... Figure 12 .

[0198] (5) The effect of pulse interference on various characteristic parameters of QPSK

[0199] The QPSK signal is subject to pulse interference, with the interference-to-signal ratio of the QPSK signal being 8 dB.

[0200] Simulate the changes in four selected characteristic parameters of a QPSK signal (signal bit rate 0.5 Mbps, sampling rate 4 MHz, signal duration 10 ms, carrier frequency 1 kHz) under pulse interference. The pulse interference is a triangular pulse with a pulse count of 4.

[0201] Because impulse interference is sudden and lasts for a period of time in the time domain, it has a significant impact on the time-domain amplitude entropy of the signal. Since the signal energy is concentrated in the time domain, it exhibits frequency broadening when transformed to the frequency domain, thus affecting the symmetry and power spectral entropy of the communication signal. It also has a significant impact on the higher-order cumulant characteristics F1 and F2 of the signal. For example... Figure 13 .

[0202] 3.8 Changes in the Disturbance Characteristic Parameters of the PSK Signal

[0203] Since the spectrum of 8PSK is a continuous spectrum without discrete spectral lines, and as discussed above, only single-tone and multi-tone interference affect the number of discrete peaks in the signal spectrum. Therefore, this section's analysis of the impact of 8PSK disturbances on characteristic parameters will no longer analyze the number of spectral peaks. Spectral symmetry is affected by linear frequency modulation (LFM) interference and pulse interference; therefore, this section's analysis of changes in 8PSK characteristic parameter values ​​due to disturbances will only analyze H... a H f With F8.

[0204] (1) The effect of single-tone interference on the characteristic parameters of 8PSK

[0205] Simulate the changes in selected characteristic parameters of an 8PSK signal (signal bit rate 0.5Mbps, sampling rate 4MHz, signal duration 10ms, carrier frequency 1KHz) under single-tone interference.

[0206] It is known that the power spectrum of an 8PSK signal has no discrete spectrum. However, after being subjected to single-tone interference, a discrete spectral line appears in its power spectrum, thus affecting the H signal. f , making H f It is smaller than when there is no interference. Furthermore, as the interference-to-signal ratio increases, H... f The signal becomes smaller. Single-tone interference alters the time-domain energy of the signal; as the interference power increases, the time-domain energy becomes more dispersed, H... a The interference-to-signal ratio increases, but it needs to be above 8 dB to be distinguishable from the interference-free state, and the fluctuation is relatively small. For example... Figure 14 .

[0207] (2) The effect of multi-tone interference on the characteristic parameters of 8PSK

[0208] Simulate the changes in selected characteristic parameters of an 8PSK signal (signal bit rate 0.5Mbps, sampling rate 4MHz, signal duration 10ms, carrier frequency 1KHz) under single-tone interference.

[0209] Multitone interference causes the received signal spectrum to consist of both a continuous spectrum and multiple discrete peaks, thus reducing the power spectral entropy. Multitone interference alters the signal's time-domain energy; as the interference power increases, the signal's time-domain energy becomes more concentrated than when there is no interference. a The fluctuation is reduced, but smaller than when there is no interference. Multi-tone interference causes large variations in F8. For example... Figure 15 .

[0210] (3) The effect of linear frequency modulation interference on the characteristic parameters of 8PSK

[0211] Simulate the changes in selected characteristic parameters of an 8PSK signal (signal code rate 0.5Mbps, sampling rate 4MHz, signal duration 10ms, carrier frequency 1KHz) under linear interference.

[0212] Linear frequency modulation (LFM) interference significantly alters the signal's spectral characteristics, resulting in large changes in power spectral entropy. The F8 value fluctuates noticeably after the signal is subjected to multi-tone interference. For example... Figure 16 .

[0213] (4) The effect of narrowband interference on the characteristic parameters of 8PSK

[0214] Simulate the changes of selected characteristic parameters of an 8PSK signal (signal code rate 0.5Mbps, sampling rate 4MHz, signal duration 10ms, carrier frequency 1KHz) under narrowband interference.

[0215] Narrowband interference significantly alters the spectral characteristics of a signal, resulting in large variations in power spectral entropy. The F8 value fluctuates noticeably after narrowband interference. While narrowband interference affects the time-domain amplitude entropy, the difference between this value and the value without interference is relatively small. Figure 17 .

[0216] (5) The effect of pulse interference on the characteristic parameters of 8PSK

[0217] Simulate the changes in selected characteristic parameters of an 8PSK signal (signal code rate 0.5Mbps, sampling rate 4MHz, signal duration 10ms, carrier frequency 1KHz) under pulse interference.

[0218] Impulse interference is characterized by its sudden onset and sustained duration in the time domain, thus significantly impacting the signal's time-domain amplitude entropy. Since the signal's energy is concentrated in the time domain, it exhibits frequency broadening when transformed to the frequency domain, affecting the symmetry and power spectral entropy of the communication signal. It also has a substantial impact on the signal's higher-order cumulant characteristic, F8. For example... Figure 18 .

[0219] The effectiveness of this invention can be further demonstrated through the following simulation experiments.

[0220] Dataset 1 is defined as the test data for BPSK, containing 200 data points of BPSK without interference and 100 data points each of the five types of interference, for a total of 700 data points. Dataset 2 is defined as the test data for QPSK, containing 200 data points of QPSK without interference and 100 data points each of the five types of interference, for a total of 700 data points. Dataset 3 is defined as the test data for 8PSK, containing 200 data points of 8PSK without interference and 100 data points each of the five types of interference, for a total of 700 data points. Dataset 4 is defined as the combination of datasets 1-3. Simulations are performed, and the simulation content and results are as follows:

[0221] Simulation signals were generated using MATLAB, with a communication signal bit rate of 0.5 Mbps, a sampling rate of 4 MHz, and a frequency offset ranging from -12 kHz to 10 kHz in 100 Hz steps. Relevant parameters for single-tone interference, multi-tone interference, linear frequency modulation interference, pulse interference, and narrowband interference were randomly selected from the table below. The multi-tone interference simulation data used four frequency points, with four values ​​randomly selected without repetition from the three frequency point parameters in the table below.

[0222] Table 3 Interference Signal Parameters

[0223]

[0224]

[0225] The test data consists of 200 data points for each modulation type without interference, with 100 data points for each modulation method subjected to five different interference signals. Data sets 1-3 are defined as test data for BPSK, QPSK, and 8PSK, respectively, and data set 4 is defined as the combination of data sets 1-3.

[0226] Simulation Experiment 1

[0227] Based on Table 2, a decision tree-based disturbed state identification algorithm was set up for each type of signal, and the PSK disturbed state identification algorithm was simulated. The accuracy of the disturbed state identification algorithm for PSK signals was verified using signal analysis of dataset 4. The results are as follows: Figure 19 As shown.

[0228] The above experiments demonstrate the stability of the selected parameters. The PSK-type communication signal interference state identification algorithm achieves an accuracy of 80% with an interference-to-signal ratio (ISR) of -5dB when the bit signal-to-noise ratio (SNR) is above 8dB, and an accuracy of over 90% when the SNR is above -5dB. Furthermore, the accuracy of interference identification increases with the increase of the SNR.

[0229] Simulation Experiment 2

[0230] Based on Table 1, a decision tree-based disturbance state identification algorithm was set up for each modulation type signal, and the performance of the decision tree for each modulation type signal was analyzed. The test data is dataset 1-3.

[0231] For the disturbance state identification rate of each modulation type of communication signal, when the bit signal-to-noise ratio (SNR) is above 8dB, the disturbance state identification accuracy of BPSK, QPSK, and 8PSK is 80% when the interference signal-to-signal ratio (ISR) is -10dB, with 8PSK achieving an accuracy of 98%. The accuracy improves to 95% when the INR is -5dB. The disturbance identification accuracy increases with increasing INR. Figure 20 As shown.

[0232] Simulation Experiment 3

[0233] An SVM (Support Vector Machine) model was trained for PSK modulation type disturbance state identification. Training data was regenerated, with the sampling rate adapted to eight times the bit rate, and the frequency offset incremented by 500Hz from -12kHz to 12.5kHz. Relevant parameters for single-tone interference, multi-tone interference, linear frequency modulation interference, impulse interference, and narrowband interference were randomly selected from the table below. The multi-tone interference simulation data used four frequency points, with four values ​​randomly selected without repetition from the four parameters in the table below. The test set consisted of four parameters.

[0234] Table 4. Interference signal parameters used for SVM training

[0235]

[0236]

[0237] The results are as follows Figure 19 As shown.

[0238] Simulation Experiment 4

[0239] For each type of modulation disturbance state identification, an SVM (Support Vector Machine) model was trained, and the training data was the same as in Experiment 3.

[0240] The test data consists of datasets 1-3. The accuracy of perturbation state identification using the SVM model on the test set is shown in the following figure:

[0241] The test set uses an SVM model to achieve the following accuracy in identifying disturbed states: Figure 21 As shown.

[0242] For the disturbance state identification rate of each modulation type of communication signal, when the bit signal-to-noise ratio is above 5dB, the disturbance state identification rate of each modulation type is higher than the accuracy rate based on the decision tree algorithm, reaching 80%. When the bit signal-to-noise ratio is above -3dB, the accuracy rate reaches 90%. When the bit signal-to-noise ratio is above -3dB, the disturbance state identification rate of each modulation type reaches over 98%.

Claims

1. A method of identifying a disturbed state of a PSK digitally modulated communication signal, characterized in that Specifically comprising the following steps: Step one, the characteristic parameters of different modulation type signals are analyzed, including the analysis of the characteristic parameters of the spectrum symmetry P, the number of spectrum discrete peaks Peak_num, the time domain amplitude entropy H a , the power spectrum entropy H f and the high order cumulant ratio F1, F2 and F8, to obtain the characteristic parameters for the disturbed state recognition; 1) analyzing the characteristic parameter of spectral symmetry: The spectral symmetry P is a reflection of whether the spectrum of the communication signal is symmetric about the carrier frequency; when the signal spectrum is symmetric, the theoretical value of the spectral symmetry P is 0; according to the PSK signal having a symmetric spectrum with the carrier frequency as the center, the spectral symmetry is maintained when it is not affected by the interference; it is concluded that when the signal is subjected to the suppressing interference and the frequency point of the interference is not at the carrier frequency of the signal, the spectral characteristics of the PSK signal spectrum change and the spectral symmetry is destroyed; 2) analyzing the number of spectral discrete peaks Peak_num, i.e. the number of signal frequency domain discrete spectrum: Firstly, the spectral data is preprocessed; then the spectral line is extracted to obtain the number of spectral discrete peaks Peak_num; the spectrum of the PSK signal has no discrete spectrum, and the number of spectral peaks Peak_num of the PSK signal is equal to 0 when it is not affected by the interference; 3) On the time-domain amplitude entropy H a Analysis: Let the received signal be r(t), sampled at a rate f s The samples are obtained as r(n), where n = 1, 2, ···, N, N is the total number of samples; the instantaneous amplitude of the signal can be expressed as a(n) = |r(n)|; the time-domain amplitude entropy of the signal is obtained using the instantaneous amplitude of the signal: 4) on the power spectrum entropy H f Analysis: After the received signal is discretely sampled, r(n) is obtained, and the power spectrum X(w) of the signal is calculated by using the periodogram method; where R(w) is the Fourier transform of r(n), and the entropy H of the power spectrum of r(n) is obtained f ; 5) analyzing the high-order cumulant ratios F1, F2 and F8 The high-order cumulant of the Gaussian random variable is zero, and the high-order cumulant of the non-Gaussian random variable is not zero; the high-order cumulants corresponding to the digital signal containing the Gaussian white noise and the digital signal without noise are theoretically the same; For a complex random process X(t) with zero mean, the high-order moment is defined as: M pq = E[X(t (p-q) X * (t) q ] The cumulant is defined as: C pq = C um {X(t),···,X(t),X * (t),···,X * (t)} where X(t) is the p-q term, X * (t) is the q term, C um is the cumulative moment, and * denotes the conjugate. The relationship between the cumulants and the moments is as follows: C 20 = M 20 C 21 = M 21 C 40 = M 40 -3M 20 2 C 42 = M 42 - |M 20 | 2 -2M 21 2 C 60 = M 60 - 15M 40 M 20 + 30M 20 3 Since the signal s(t) and the Gaussian white noise n(t) are independent, according to the properties of the cumulants, it can be obtained that: C um (r(t)) = C um (s(t)) + C um (n(t)) And the zero-mean Gaussian white noise is greater than the value of the second-order cumulant, which is zero, so the above formula can be expressed as: C um (r(t)) = C um (s(t)); The theoretical model of the received signal is: where E is the energy of the modulated symbol, a k is the energy normalized digital symbol sequence, T s is the symbol period, ω c is the carrier frequency, θ is the initial phase of the carrier, where the symbol values occur with equal probability; n(t) is a zero-mean Gaussian white noise; after the phase, carrier frequency, and timing of the received signal are synchronized, the received signal is down-converted, and after timing sampling, the theoretical high-order cumulants of PSK signals are as shown in Table 1: Table 1 Theoretical high-order cumulants of the PSK signal The high-order cumulants of the signals in Table 1 are all related to E, and the high-order cumulant ratios of the signals of the same modulation type exist differences; the characteristic parameters related to the high-order cumulant ratios are constructed to eliminate the influence of E on the high-order cumulant ratios of the signals of the same modulation type: wherein Step two, according to the characteristic parameters of the spectral symmetry, the number of spectral discrete peaks, the time domain amplitude entropy, the power spectrum entropy and the high-order cumulant ratio, the stable range of the PSK characteristic parameters without interference is determined to determine whether the signal is disturbed.

2. The method of claim 1, wherein the method is characterized by: The characteristic parameter of the spectral symmetry in step one is analyzed as follows: The spectral symmetry P is a reflection of whether the spectrum of the communication signal is symmetric about the carrier frequency In the formula, where P U , P L represent the spectrum of the signal upper and lower sidebands, respectively, the upper sideband being divided by the carrier frequency; When the signal spectrum is symmetric, the theoretical value of the spectral symmetry P is 0; according to the PSK signal having a symmetric spectrum with the carrier frequency as the center, the spectral symmetry is maintained when it is not affected by the interference; it is concluded that when the signal is subjected to the suppressing interference and the frequency point of the interference is not at the carrier frequency of the signal, the spectral characteristics of the PSK signal spectrum change and the spectral symmetry is destroyed.

3. A method of identifying a disturbed state of a PSK digitally modulated communication signal according to claim 2, characterized in that, The number of spectral discrete peaks Peak_num, i.e. the number of signal frequency domain discrete spectrum, in step one is analyzed as follows: First, the spectrum data is preprocessed: using {R(f i For each frequency point f of the group i = 1, 2, ..., N, ... i Spectral value at f i-A ~f i-B and f i+A ~f i+B The spectrum R'(f) is obtained by taking the ratio of the mean of the two m maximum values: Then the spectral line is extracted to obtain the number of spectral discrete peaks Peak_num; the spectrum of the PSK signal has no discrete spectrum, and the number of spectral peaks Peak_num of the PSK signal is equal to 0 when it is not affected by the interference.

4. The method of claim 1, wherein the method is characterized by: The step one in the time domain amplitude entropy H a The analysis is as follows: Let the received signal be r(t), sampled at a rate f s The samples are obtained as r(n), where n = 1, 2, ···, N, N is the total number of samples; the instantaneous amplitude of the signal can be represented as a(n) = |r(n)|; the time-domain amplitude entropy of the signal is obtained using the instantaneous amplitude of the signal: The instantaneous amplitude a(n) is normalized The time-domain amplitude entropy of the signal is wherein is the proportion of each amplitude energy to the total energy; Time domain amplitude entropy H a The time domain amplitude entropy is used to characterize the dispersion degree of the amplitude value energy of the digital signal. The greater the time domain amplitude entropy is, the more dispersed the amplitude value energy is. The smaller the time domain amplitude entropy is, the more concentrated the amplitude value energy is. The PSK signal transmits information according to the phase change in the modulation process. When the amplitude value energy is uniformly distributed in the sampling interval and is not affected by the interference, the value of the amplitude entropy is stable.

5. The method of claim 1, wherein the method is characterized by: The step one power spectrum entropy H f Analysis as follows: The received signal is discretely sampled as r(n), and the power spectrum X(w) of the signal is calculated by using the periodogram method: In the formula, R(w) is the Fourier transform of r(n), and the power spectrum entropy of r(n) is: Power spectral entropy H f The power spectral entropy characterizes the degree of energy dispersion in a digital signal, reflecting the number of power spectral lines. A higher power spectral entropy indicates a greater number of spectral lines, while a lower entropy indicates fewer lines. For PSK-type signals, whose power spectrum only has continuous spectral components, their energy is relatively dispersed. When unaffected by interference, the H of PSK... f It is relatively stable, but if the signal is interfered with, its power spectrum energy dispersion will be disrupted, i.e., H f Fluctuations occurred.

6. The method of claim 1, wherein the method further comprises: determining the number of the PSK digital modulation communication signals; and determining the number of the PSK digital modulation communication signals in the disturbed state. The step two determines the stable range of the PSK characteristic parameters without interference according to the spectral symmetry, the number of spectral discrete peaks, the time-domain amplitude entropy, the power spectrum entropy, and the characteristic parameters of the high-order cumulant ratio, and determines whether the signal is disturbed or not, and the specific steps are as follows: 1) Determine the stable range of the PSK signal characteristic parameters without interference: The stable ranges of the characteristic parameters of BPSK, QPSK and 8PSK signals at no interference are calculated respectively when the bit signal-to-noise ratio is 8-20 dB, the characteristic parameter is calculated for not less than 50 times at each bit signal-to-noise ratio, and the average value is taken as the characteristic parameter value at the bit signal-to-noise ratio; the symbols are randomly generated when the BPSK, QPSK and 8PSK baseband signals are generated, and the oversampling rate of the baseband signal is 8; the characteristic parameter stable ranges of the spectral symmetry P, the spectral discrete peak number Peak_num, the time domain amplitude entropy H a , the power spectrum entropy H f and the high-order cumulant ratio F1, F2 and F8 within the bit signal-to-noise ratio of 8-20 dB are obtained, as shown in Table 2 Table 2 Stable range of PSK signal characteristic parameters without interference 2) Take the stable range of the PSK characteristic parameters without interference in Table 2 as the basis for judging whether the signal is disturbed or not, select the spectral symmetry, the number of spectral discrete peaks, the time-domain amplitude entropy, and the power spectrum entropy as the characteristic parameters for the disturbed judgment for the PSK signal, and increase the selection of F1 as the characteristic parameter for the disturbed judgment for the BPSK modulation type on the basis of the selection of the spectral symmetry, the number of spectral discrete peaks, the time-domain amplitude entropy, and the power spectrum entropy, increase the selection of F1 and F2 as the characteristic parameters for the disturbed judgment for the QPSK on the basis of the selection of the spectral symmetry, the number of spectral discrete peaks, the time-domain amplitude entropy, and the power spectrum entropy, and increase the selection of F1 and F8 as the characteristic parameters for the disturbed judgment for the 8PSK on the basis of the selection of the spectral symmetry, the number of spectral discrete peaks, the time-domain amplitude entropy, and the power spectrum entropy; when one or more characteristic parameters exceed the range, it is determined that the signal is disturbed.

7. A system for identifying a disturbed state of a PSK digitally modulated communication signal based on the identification method according to claims 1 to 6, characterized in that It includes: An input module for providing prior conditions required by the system, i.e., the signal modulation mode; Input the sampling data of the received signal; A characteristic parameter calculation module for calculating the characteristic parameters of the received signal for subsequent disturbed state recognition; An output module for using the modulation mode of the input signal to call the corresponding disturbed state recognition algorithm and output the disturbed state of the signal.

8. A disturbed state recognition device for PSK digitally modulated communication signals based on the recognition method according to claims 1 to 6, characterized in that It includes: A memory for storing a computer program; A processor for executing the computer program to realize the disturbed state recognition method of the PSK digital modulation communication signal according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program can recognize the disturbed state of the PSK digital modulation communication signal according to any one of the methods of claims 1 to 6 when executed by the processor.

Images