Frequency hopping signal judgment threshold value self-adaptive adjustment method

Abstract

The invention relates to a frequency hopping signal judgment threshold value self-adaptive adjustment method, which comprises the following steps of: acquiring a power spectrum of a signal to be detected, and calculating to obtain an average power spectrum and a maximum retention spectrum; the average power spectrum is shifted to be close to the maximum retention spectrum bottom noise, and the shifted average power spectrum serves as an initial threshold value; and judging whether a frequency hopping signal exists or not, and if the frequency hopping signal exists, selecting a corresponding adjustment coefficient according to the size of the signal-to-noise ratio to dynamically adjust the threshold value of the frequency hopping frequency point. The judgment threshold value generated by the method comprehensively considers various factors in a complex electromagnetic environment, is low in calculation complexity, has wide applicability, and improves the accuracy and reliability of frequency hopping signal detection.

Description

Technical Field

[0001] The present invention relates to the technical field of frequency-hopping signal detection, and particularly to a method for adaptively adjusting the decision threshold value of frequency-hopping signals. Background Art

[0002] Frequency-hopping technology is a spread-spectrum communication technology that transmits information by quickly switching between multiple frequencies. It has the advantages of strong anti-interference ability, good concealment, and high reliability, and is widely used in military communications, UAV control, wireless sensor networks and other fields. However, in practical applications, the detection of frequency-hopping signals faces many challenges. First, the modern electromagnetic environment is complex and changeable, with a variety of interference sources, such as other wireless communication systems, radars, electronic warfare equipment, etc. These interference sources may generate signals similar to frequency-hopping signals, increasing the detection difficulty. Second, the signal intensity of frequency-hopping signals may vary greatly at different frequencies, especially in the case of long-distance transmission and multipath propagation. Traditional fixed-threshold detection methods are difficult to adapt to this fluctuating environment, and are prone to false alarms and missed detections. Traditional detection methods mainly include fixed-threshold detection, energy detection, and correlation detection. The fixed-threshold detection method determines whether there is a frequency-hopping signal by setting a fixed detection threshold, but this method is difficult to adapt to different signal environments and is prone to false alarms and missed reports. The energy detection method determines whether there is a frequency-hopping signal by calculating the energy of the signal. Although it is simple and effective, the detection performance will decrease significantly in a low signal-to-noise ratio environment. The correlation detection method detects frequency-hopping signals by performing correlation operations with known frequency-hopping sequences, but it has high requirements for prior knowledge of frequency-hopping sequences and high computational complexity.

[0003] In order to overcome the limitations of traditional detection methods, an adaptive threshold detection method has been proposed. The core idea of adaptive threshold detection is to dynamically adjust the threshold value during the detection process to adapt to different signal environments. Existing adaptive threshold detection methods have problems such as a single adjustment strategy, high computational complexity, and limited applicability. Therefore, an adaptive threshold algorithm for frequency-hopping signal detection that can comprehensively consider multiple factors, has low computational complexity, and has wide applicability is needed. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method for adaptively adjusting the decision threshold value of frequency-hopping signals, which can comprehensively consider multiple factors in a complex electromagnetic environment, has low computational complexity, and has wide applicability.

[0005] The technical solution adopted by the present invention to solve its technical problems is: to provide a method for adaptively adjusting the decision threshold value of frequency-hopping signals, including the following steps:

[0006] Obtain the power spectrum of the signal to be measured, and calculate the average power spectrum and the maximum hold spectrum;

[0007] Shift the average power spectrum to near the maximum hold spectral floor noise, and use the shifted average power spectrum as the initial threshold value;

[0008] Determine whether there is a frequency hopping signal. If there is a frequency hopping signal, select a corresponding adjustment coefficient according to the signal-to-noise ratio to dynamically adjust the threshold value of the frequency hopping frequency point.

[0009] Further, the step of selecting a corresponding adjustment coefficient according to the signal-to-noise ratio to dynamically adjust the threshold value of the frequency hopping frequency point includes:

[0010] Increase the threshold value for the first frequency hopping frequency point with a signal-to-noise ratio greater than the first threshold;

[0011] Decrease the threshold value for the second frequency hopping frequency point with a signal-to-noise ratio less than the second threshold and greater than or equal to the third threshold;

[0012] Increase the threshold value for the third frequency hopping frequency point with a signal-to-noise ratio less than the third threshold and greater than zero, and the adjustment amount of this threshold value is less than the adjustment amount of the threshold value of the first frequency hopping frequency point.

[0013] Further, the threshold value Th hop (k) is expressed as

[0014]

[0015] where k is the frequency point, Th x (k) is the initial threshold value, ΔTh x is the threshold correction value, S max (k) is the maximum hold spectrum, SNR(k) is the signal-to-noise ratio, Th snr1 、Th snr2 and Th snr3 are the first threshold, the second threshold and the third threshold respectively, and β1, β2 and β3 are all set parameters.

[0016] Further, the threshold correction value ΔTh x is the weighted sum of the mean and variance of the maximum hold spectrum corresponding to the frequency points with a signal-to-noise ratio less than zero.

[0017] Further, it also includes the step of performing a sliding average based on the frequency point on the dynamically adjusted threshold value.

[0018] Further, the threshold value autoTh x (k) is expressed as

[0019]

[0020] where 2n + 1 is the sliding window length and i is the frequency point within the sliding window.

[0021] Further, the offset Th of the average power spectrum to near the bottom noise of the maximum hold spectrum offset is calculated according to the following formula

[0022]

[0023] where k is the frequency point, is the average power spectrum, and S max (k) is the maximum hold spectrum, and α is the correction factor.

[0024] Further, it also includes the step of calculating the signal-to-noise ratio of each frequency point by calculating the difference between the maximum hold spectrum and the initial threshold value.

[0025] Further, the determination of whether there is a frequency-hopping signal includes:

[0026] Using the dynamically adjusted threshold value to determine whether there is a frequency-hopping signal;

[0027] If there is no dynamically adjusted threshold value, the frequency points with a signal-to-noise ratio greater than zero are used as the frequency-hopping frequency points.

[0028] Further, the obtaining of the signal power spectrum includes:

[0029] Obtaining the baseband sampled signal and performing STFT transformation to obtain discrete frequency domain data;

[0030] Taking the square of the modulus of the discrete frequency domain data and dividing it by the STFT window length to obtain the signal power spectrum.

[0031] Further, the calculation of the average power spectrum and the maximum hold spectrum includes:

[0032] Performing equally spaced extraction on the signal power spectrum;

[0033] Based on the extracted power spectrum, calculating the average of each frequency point in the time dimension to obtain the average power spectrum, and calculating the maximum value of each frequency point in the time dimension to obtain the maximum hold spectrum.

[0034] Beneficial effects

[0035] Due to the adoption of the above technical solutions, compared with the prior art, the present invention has the following advantages and positive effects: By shifting the average power spectrum curve upward to near the bottom noise of the maximum hold spectrum and using the shifted average power spectrum as the initial threshold value, the present invention can make the instantaneous power spectrum of the frequency-hopping signal cross the threshold value, thereby achieving the purpose of detecting the frequency-hopping signal and measuring the signal-to-noise ratio, and reducing the subsequent calculation complexity; at the same time, by using different adjustment coefficients at different signal-to-noise ratios to dynamically adjust the decision threshold near the position of the frequency-hopping signal, the detection performance of the frequency-hopping signal can be effectively optimized to cope with the complex and changeable signal environment, ensuring accurate and reliable detection of the frequency-hopping signal under various conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 is a flowchart of an embodiment of the present invention;

[0037] Figure 2 is a schematic diagram of adaptive threshold adjustment of the present invention embodiment at a signal-to-noise ratio of 8 dB;

[0038] Figure 3 is a schematic diagram of adaptive threshold adjustment of the present invention embodiment at a signal-to-noise ratio of 25 dB;

[0039] Figure 4 is a schematic diagram of adaptive threshold adjustment of the present invention embodiment in the application of an unmanned aerial vehicle. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] The following further elaborates the present invention in conjunction with specific embodiments. It should be understood that these embodiments are only used to illustrate the present invention and not to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the appended claims of this application.

[0041] The embodiment of the present invention relates to a method for adaptively adjusting the decision threshold value of a frequency-hopping signal, as Figure 1 shown, including the following steps:

[0042] (1) Signal acquisition step

[0043] Use a suitable signal acquisition device to collect signals in the wireless channel in real time. The signal acquisition device includes, but is not limited to, components such as radio frequency antennas and receivers, which can receive and process signals in multiple frequency bands from the wireless channel. The collected signals are subjected to necessary front-end processing, such as amplification, filtering, etc., to ensure the signal quality for subsequent processing.

[0044] More specifically, the intermediate frequency signal is down-converted and filtered to obtain a baseband sampling signal, specifically represented as x(n), n = 0, 1, ….

[0045] The down-conversion and filtering processing here can be implemented by using conventional down-conversion circuits and filtering circuits in the prior art. The frequency conversion parameters of the down-conversion and the cut-off frequency of the filtering can be flexibly set according to the actual application scenario and the characteristics of the wireless channel targeted.

[0046] In some embodiments, the signal acquisition device may further include a signal preprocessing module, which is capable of performing additional preprocessing operations on the acquired signals before entering the down-conversion filtering process, such as removing the DC component, performing preliminary noise suppression, etc. These preprocessing operations can be implemented using various signal processing techniques and algorithms known in the prior art, and their specific processing parameters and algorithms can be adjusted according to the characteristics of the actually acquired signals.

[0047] (2) Steps for calculating the time-frequency diagram

[0048] The discrete STFT transform is used to calculate the time-frequency diagram of the signal. The discrete baseband signal x(n) is subjected to the discrete STFT transform to obtain

[0049]

[0050] w(i) represents the window function. The window function can be selected from but not limited to common window function types such as the Hanning window and the rectangular window, and its window length, shape parameters, etc. can be adjusted according to specific requirements. N is the window length, that is, the length of each frame of data participating in the calculation. The value of the window length can be determined comprehensively according to the required time resolution and frequency resolution. For example, when detecting a rapidly changing signal, the window length can be appropriately shortened to improve the time resolution, while when focusing on frequency details, the window length can be appropriately increased to improve the frequency resolution. n is the sampling time point, and k is the discrete frequency point. In actual situations, considering the computational complexity, usually let n = ld, where d is the step length each time, then there is

[0051]

[0052] The value of the step length d can be flexibly set according to the characteristics of the signal and the requirements of subsequent processing. For example, when it is desired to analyze the signal more finely in the time dimension, the value of d can be reduced. The selection of the length of the window function w(i) affects both the time and frequency resolutions. If the selected window is too short, the time resolution is very high, but the frequency resolution is very low; on the contrary, if the selected window is too long, the signal has a very high frequency resolution, but the time resolution is reduced. In order to balance both the time resolution and the frequency resolution, usually the baseband signal x(n) needs to be overlapped, usually half of the length N of each frame, that is, d = N / 2, but other overlapping ratios can also be used under certain conditions according to the actual situation, such as d = N / 4, 3N / 4, etc.

[0053] The square of the modulus of the data after the short-time Fourier transform is divided by the window length N to obtain the signal power spectrum

[0054]

[0055] The signal power spectrum is more friendly than the short-time Fourier transform in terms of anti-noise performance, which makes the focusing performance of the frequency-hopping signal in the time-frequency domain better. The modulus, square, and division operations involved in the power spectrum calculation are all performed according to the conventional mathematical operation rules, and appropriate numerical calculation methods and software tools can be used during the calculation to ensure the accuracy and efficiency of the calculation.

[0056] For the result after discrete STFT transformation, in addition to calculating the signal power spectrum, other forms of subsequent processing can be further performed on it, such as calculating the phase spectrum of the signal, performing energy normalization processing, etc. These subsequent processing operations can be selected and implemented according to specific analysis requirements and subsequent application scenarios, and their processing methods and parameters can be adjusted according to the actual situation.

[0057] (III) Steps for extracting the power spectrum

[0058] Extract the power spectrum at equal intervals to reduce the computational complexity. Let the total number of frames of the input power spectrum be, and extract at equal interval D. The total number of frames after extraction is

[0059]

[0060] Denotes rounding down. The extracted power spectrum is denoted as

[0061] S xD (m,k) = SPEC x (mD,k), m = 0, 1, …, M

[0062] The value of the interval D here can be determined comprehensively according to factors such as specific computing resources, requirements for real-time performance, and the complexity of the signal itself. For example, in the case of limited computing resources and high requirements for real-time performance, the value of D can be appropriately increased to reduce the computational complexity faster.

[0063] In addition to using the equal-interval extraction method to reduce the computational complexity, other extraction methods can also be used, such as unequal-interval extraction, selective extraction based on specific rules, etc. The specific implementation methods and parameters of these extraction methods can be determined comprehensively according to factors such as specific signal characteristics, computing resources, and requirements for real-time performance. When using these extraction methods, it is also necessary to record and perform subsequent processing on the extracted power spectrum to ensure the accurate detection of frequency-hopping signals.

[0064] (IV) Steps for calculating the average power spectrum and the maximum hold spectrum

[0065] Calculate the average power spectrum and the maximum hold spectrum of the signal to estimate the initial decision threshold, and estimate the initial threshold to calculate the signal-to-noise ratio of the frequency-hopping signal, thereby determining whether there is a frequency-hopping signal.

[0066] The average power spectrum is the average of each frequency point in the time dimension, that is

[0067]

[0068] The maximum hold spectrum is the maximum value of each frequency point in the time dimension and can be expressed as

[0069] S max (k) = max m (S xD (m, k))

[0070] By adjusting the decimation factor D, the computational complexity can be effectively reduced. In addition, the impact of instantaneous interference on the average power spectrum and the maximum hold spectrum can also be reduced.

[0071] The calculation process of the average power spectrum is carried out according to the conventional rules of summation and division operations to ensure the accuracy of the calculation results. For the calculation of the maximum hold spectrum, the decimated power spectrum can be traversed, and the maximum value that appears at each discrete frequency point can be recorded as the maximum hold spectrum value at that frequency point. The specific calculation method can be implemented using a suitable algorithm, such as a loop comparison algorithm, etc.

[0072] (V) Steps for estimating the initial threshold and signal-to-noise ratio

[0073] Based on the calculated average power spectrum and maximum hold spectrum, a suitable estimation method is used to estimate the initial decision threshold. The specific estimation methods can include, but are not limited to, estimation methods based on statistical characteristics, estimation methods based on empirical formulas, etc.

[0074] Since the fixed-frequency signal and the noise floor envelope change slowly over time, the instantaneous power spectrum is highly consistent with the average power spectrum. The average power spectrum of the frequency-hopping signal is averaged in the time dimension, so the energy density is much lower than the energy density of the instantaneous power spectrum. Therefore, in this embodiment, the average power spectrum curve is shifted upward to near the noise floor of the maximum hold spectrum, and the shifted average power spectrum is used as the initial threshold value, so that the instantaneous power spectrum of the frequency-hopping signal can cross the threshold value, thereby achieving the purpose of detecting and measuring the signal-to-noise ratio of the frequency-hopping signal. The offset value can be estimated according to the following formula:

[0075]

[0076] The first term in the formula is the main bias amount, which represents the average distance from the average power spectrum to the noise floor of the maximum hold spectrum. The second term is the threshold offset distance correction parameter, α is the correction factor, and generally 0.25 is taken in engineering. Its purpose is to make the noise floor not cross the threshold value as much as possible, while the power spectrum of the frequency-hopping signal is basically not affected. In summary, the initial threshold can be expressed as

[0077]

[0078] According to the estimated initial decision threshold, combined with relevant information such as the signal power spectrum, calculate the signal-to-noise ratio (SNR) of the frequency-hopping signal according to the conventional SNR calculation method. For example, the SNR can be determined by calculating the ratio of the signal power to the noise power. Then, compare the calculated SNR with the preset threshold. If the SNR is greater than the preset threshold, it is determined that there is a frequency-hopping signal; if the SNR is less than or equal to the preset threshold, it is determined that there is no frequency-hopping signal. The preset threshold here can be flexibly set according to factors such as the specific application scenario and the accuracy requirement for frequency-hopping signal detection. For example, in a military communication detection scenario with high detection accuracy requirements, the preset threshold can be set relatively high.

[0079] This embodiment simply estimates the SNR of each frequency of the frequency-hopping signal based on the maximum hold spectrum and the initial threshold value. In the logarithmic representation, the difference between the two is the SNR, denoted as

[0080] SNR(k) = S max (k) - Th x (k)

[0081] It can be seen from the above formula that the SNR SNR(k) is a function of the frequency point k. When SNR(i) > 0, this frequency point is a frequency hop; when SNR(i) ≤ 0, this frequency point is the background noise or a fixed frequency. Therefore, it can be determined whether there is a frequency hop in the current data according to whether there is a value greater than 0 in SNR(k). In engineering, to ensure the reliability of the judgment, it is usually necessary to count the number N of values greater than 0 in SNR(k) h , when N h is greater than the set threshold can it be considered that there is a frequency hop to avoid false detection.

[0082] For the estimation method of the initial decision threshold and the calculation method of the SNR, in addition to the conventional methods mentioned above, other advanced methods can also be used, such as the estimation method based on machine learning and the calculation method based on deep learning. These advanced methods can be trained using training data to improve the accuracy and adaptability of the estimation and calculation. And when using these advanced methods, it is necessary to reasonably arrange and manage the collection, collation, and training process of the training data to ensure that satisfactory detection results can be obtained.

[0083] (6) Steps for optimizing the frequency-hopping threshold based on the SNR

[0084] After determining the existence of a frequency-hopping signal, the frequency-hopping threshold is further optimized and adjusted according to the calculated signal-to-noise ratio SNR(k). When SNR(i) adopts different adjustment coefficients under different signal-to-noise ratios, at high signal-to-noise ratios, noise passing through the decision threshold should be suppressed as much as possible with the primary purpose of detecting high-quality frequency hopping. Therefore, the threshold should be increased upward near the frequency-hopping frequency point; at low signal-to-noise ratios, since the quality of the frequency-hopping signal is poor, the detection of the frequency-hopping signal should be prioritized as much as possible, and some threshold quality can be sacrificed to obtain better frequency-hopping detection. Therefore, the threshold should be decreased near the frequency-hopping frequency point; when the signal-to-noise ratio is extremely low, since some background noise may cross the threshold value, causing false detection, in this case, the threshold value should be slightly increased to avoid false detection.

[0085] More specifically, if the signal-to-noise ratio is higher than a first preset value (which can be preset according to the actual application scenario and system performance requirements), the frequency-hopping threshold is decreased to improve the detection sensitivity to weak frequency-hopping signals. Specifically, the threshold can be adjusted according to a preset decrease ratio (such as decreasing the current threshold by 10%, 20%, etc., which can be adjusted according to the actual situation); if the signal-to-noise ratio is lower than a second preset value (which can also be preset), the frequency-hopping threshold is increased to reduce the possibility of false judgment, and the threshold can be adjusted according to a preset increase ratio (such as increasing the current threshold by 15%, 25%, etc., which can be adjusted).

[0086] In this embodiment, various factors are comprehensively considered for dynamic adjustment, and the adjusted threshold value can be expressed as

[0087]

[0088] In the formula, ΔTh x is the final threshold correction value, which is mainly used to filter out false detections caused by large noise. Usually, it is obtained by calculating the mean and variance of {S max (k)|SNR(k)<0} and weighting them, that is:

[0089] ΔTh x = ω1E({S max (k)|SNR(k)<0}) + ω2Var({S max (k)|SNR(k)<0})

[0090] After completing the local adjustment of the threshold, the overall fluctuation of the threshold value is relatively large, and the threshold needs to be further optimized to facilitate subsequent frequency-hopping signal detection. In engineering, to save calculation time, a moving average is usually performed on k in the frequency domain. Assuming the moving window length is 2n + 1, the output threshold value is expressed as:

[0091]

[0092] After optimizing and adjusting the frequency hopping threshold, the optimized threshold is used again to detect and judge the frequency hopping signals in the subsequently collected signals, so as to continuously improve the detection accuracy and adaptability.

[0093] The following further illustrates this embodiment by constructing specific embodiments.

[0094] The frequency hopping signal is generated using simulation data, with a sampling rate of 25.6 MHz, 64 frequency set numbers, a channel interval of 200 KHz, and a hopping speed of 1000 hop / s. Fixed-frequency interference signals are added at positions -8 MHz and 8 MHz in frequency respectively. The Hamming window is used for the time-frequency diagram calculation, with a window length N = 4096 and an overlapping point number d = 2048. The adaptive thresholds for the frequency hopping signals at signal-to-noise ratios of 8 dB and 25 dB are calculated respectively, and the results are as Figure 2 and 3 shown.

[0095] The results show that the frequency hopping adaptive threshold can automatically adjust the threshold position according to the quality of the frequency hopping signal-to-noise ratio. At low signal-to-noise ratios, the threshold drops near the frequency hopping signal to detect the instantaneous frequency hopping signal to the maximum extent. Although there will be some noise crossing the threshold, since the noise crossing the threshold is uniformly distributed on the time-frequency diagram, it can be processed by morphological methods subsequently; at high signal-to-noise ratios, the threshold rises near the frequency hopping signal, thereby effectively suppressing the noise from passing through the threshold value and further improving the signal quality. In addition, it can be significantly found from the experimental results that the maximum hold spectrum of the fixed-frequency signal will not cross the threshold value. Even when the background noise is colored noise, this algorithm can still effectively suppress it. Currently, this algorithm can still achieve effective detection at a signal-to-noise ratio of 5 dB, and its detection range has significant advantages compared with the fixed-threshold detection method.

[0096] Figure 4 To actually collect the frequency hopping signal of a certain UAV, it can be seen from the figure that although there are obvious unevenness and other complex situations in the signal background noise, and there are complex pulses, fixed frequencies, etc. in the space, the threshold of this algorithm can still excellently adapt to the detection environment. Although there are some noise pulses that can cross the threshold value, in the subsequent sorting algorithm, since their residence time is shorter than that of the real frequency hopping signal, they will be judged as interference noise and effectively filtered, further verifying the effectiveness and stability of this algorithm in actual application scenarios.

Claims

1. A method for adaptively adjusting the decision threshold value of a frequency-hopping signal, characterized in that It includes the following steps: Obtain the power spectrum of the signal to be measured, and calculate the average power spectrum and the maximum hold spectrum; Shift the average power spectrum to near the noise floor of the maximum hold spectrum, and use the shifted average power spectrum as the initial threshold value; Determine whether there is a frequency hopping signal. If there is a frequency hopping signal, select the corresponding adjustment coefficient according to the signal-to-noise ratio to dynamically adjust the threshold value of the frequency hopping frequency point.

2. The method according to claim 1, characterized in that, The step of selecting the corresponding adjustment coefficient according to the signal-to-noise ratio to dynamically adjust the threshold value of the frequency hopping frequency point includes: Increase the threshold value for the first frequency hopping frequency point with a signal-to-noise ratio greater than the first threshold; Decrease the threshold value for the second frequency hopping frequency point with a signal-to-noise ratio less than the second threshold and greater than or equal to the third threshold; Increase the threshold value for the third frequency hopping frequency point with a signal-to-noise ratio less than the third threshold and greater than zero, and the adjustment amount of this threshold value is less than the adjustment amount of the threshold value of the first frequency hopping frequency point.

3. The method according to claim 2, wherein The threshold value Th after dynamic adjustment hop (k) is expressed as Among them, k is the frequency point, Th x (k) is the initial threshold value, ΔTh x is the threshold correction value, S max (k) is the maximum hold spectrum, SNR(k) is the signal-to-noise ratio, Th snr1 、Th snr2 and Th snr3 are the first threshold, the second threshold and the third threshold respectively, and β1, β2 and β3 are all set parameters.

4. The method according to claim 3, characterized in that, Threshold correction value ΔTh x It is the weighted sum of the mean and variance of the maximum hold spectrum corresponding to the frequency points where the signal-to-noise ratio is less than zero.

5. The method according to claim 2, wherein It also includes the step of performing a sliding average based on the frequency point on the dynamically adjusted threshold value.

6. The method according to claim 5, characterized in that, Threshold value autoTh after moving average x (k) is expressed as Among them, 2n + 1 is the sliding window length, and i is the frequency point within the sliding window.

7. The method according to claim 1, wherein The offset Th of the average power spectrum to the maximum-hold spectrum near the noise floor is calculated according to the following formula offset , and is calculated according to the following formula where k is the frequency point, is the average power spectrum, and S max (k) is the maximum hold spectrum, and α is the correction factor.

8. The method according to claim 1, characterized in that, It also includes the step of calculating the signal-to-noise ratio of each frequency point by calculating the difference between the maximum hold spectrum and the initial threshold value.

9. The method according to claim 8, wherein The determination of whether there is a frequency hopping signal includes: Use the dynamically adjusted threshold value to determine whether there is a frequency hopping signal; If there is no dynamically adjusted threshold value, use the frequency points with a signal-to-noise ratio greater than zero as the frequency hopping frequency points.

10. The method according to claim 1, characterized in that, The calculation of the average power spectrum and the maximum hold spectrum includes: Perform equally spaced sampling on the signal power spectrum; Based on the sampled power spectrum, calculate the average of each frequency point in the time dimension to obtain the average power spectrum, and calculate the maximum value of each frequency point in the time dimension to obtain the maximum hold spectrum.

Images