Frequency hopping signal detection method for filter

Abstract

The invention discloses a frequency hopping signal detection method for a filter, and relates to the technical field of filters, and the method comprises the steps: S1, setting an expected frequency hopping range of a target frequency hopping signal, and dividing the expected frequency hopping range of the target frequency hopping signal into a plurality of sub-frequency bands; s2, converting the target frequency hopping signal to a frequency domain by using PSK phase modulation, and then performing phase modulation to generate a modulation signal; s3, performing VCO frequency modulation on the modulation signal of each sub-band by using a phase-locked loop frequency tracking module; s4, all sub-frequency bands output after VCO frequency modulation are subjected to PSK phase demodulation, modulation signals are converted to a time domain, and the modulation signals are marked as corrected frequency hopping signals. Compared with the prior art, the target frequency hopping signal is converted to the frequency domain for modulation optimization through PSK phase modulation, frequency synchronization is carried out on the modulation signal of each sub-band through the phase-locked loop technology, and the method has the advantages that phase drift caused by frequency mismatching can be avoided, and higher-precision signal detection can be achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of filters, and in particular to a frequency hopping signal detection method for filters. Background Art

[0002] In a filter frequency hopping communication system, the filter needs to respond to frequency changes and quickly adjust between multiple frequency bands. When performing high-speed frequency hopping or complex multi-band frequency hopping, there is a high probability that the filter will have a random time response delay problem. Time response delay is the time required for the filter to adapt to the new frequency after receiving the signal. The response delay makes it take the filter longer to identify and lock the frequency range of the signal. This delay makes it impossible for the filter to adjust its operating frequency in time when the signal frequency jumps, and it may miss the key part of the signal, resulting in signal capture failure or incomplete capture.

[0003] The time response delay of the filter directly affects the capture rate of the system. The capture rate refers to the ability of the system to accurately identify and lock onto the signal in a sufficiently short time. When the response delay of the filter is large, the time required to capture the signal increases, resulting in a longer adaptation process for the system to the signal, increasing the chance of signal loss or demodulation errors. For example, the filter needs to adjust its frequency band to adapt to the jump or frequency change of the signal, but the delay causes the filter adjustment process to lag behind the change of the signal. Over time, this lag effect may cause the captured signal to no longer be accurate, causing the signal to have a phase error, and the filter may not be able to accurately lock onto the signal within the correct time and frequency range. When the filter cannot accurately lock onto the signal, the signal quality and demodulation accuracy will be reduced, which will cause the filter to significantly reduce the accuracy of signal detection during the frequency hopping process. Summary of the invention

[0004] The present invention provides a frequency hopping signal detection method for a filter, which solves the problem that the existing frequency hopping filter is prone to signal phase error due to time response delay during high-speed frequency hopping or complex multi-band frequency hopping, thereby reducing the signal detection accuracy.

[0005] The present invention is achieved through the following technical solutions: A frequency hopping signal detection method for a filter, the method comprising: Step S1: Select a target filter and a target frequency hopping signal for frequency hopping, collect historical frequency hopping data of the target frequency hopping signal and use it to set an expected frequency hopping range, and divide the expected frequency hopping range of the target frequency hopping signal into a plurality of sub-frequency bands; Step S2: starting the frequency hopping process of the target filter, using PSK phase modulation to convert the target frequency hopping signal to the frequency domain and then performing phase modulation along each sub-band to generate a modulated signal, and maintaining smooth transition phase control between adjacent phase modulations; Step S3: construct a phase-locked loop frequency tracking module, and use the phase-locked loop frequency tracking module to perform VCO frequency modulation on the modulation signal of each sub-band to complete the frequency synchronization of the modulation signal in the phase modulation process of each sub-band; Step S4: convert all sub-bands output after VCO frequency modulation into time domain through PSK phase demodulation and mark them as modified frequency hopping signals, apply the modified frequency hopping signals to the target filter and re-implement the detection of the frequency hopping signals.

[0006] When the response delay of the filter is large, the time required to capture the signal increases, resulting in a prolonged adaptation process of the system to the signal, increasing the probability of signal loss or demodulation errors. As time goes by, this lag effect may cause the captured signal to no longer be accurate, causing the signal to produce a phase error, and the filter may not be able to accurately lock the signal within the correct time and frequency range. When the filter cannot accurately lock the signal, the quality of the signal and the demodulation accuracy will be reduced, which will lead to a significant reduction in the signal detection accuracy of the filter during the frequency hopping process. Based on this, the present invention provides a frequency hopping signal detection method for a filter, which solves the problem that the existing frequency hopping filter is prone to signal phase errors due to time response delays during high-speed frequency hopping or complex multi-band frequency hopping, thereby reducing the signal detection accuracy.

[0007] Further, the process of modulating each sub-frequency band using PSK phase modulation includes: the historical frequency hopping data includes a signal-to-noise ratio mean value and a frequency change curve corresponding to the xy coordinate axis representing time-frequency, the total amount of the absolute value of the frequency change is collected for each sub-frequency band in the frequency change curve, and the total frequency hopping duration is obtained based on the time axis on the horizontal axis of the frequency change curve; the average frequency hopping rate of the target frequency hopping signal is calculated in each sub-frequency band according to the total frequency hopping duration and the total amount of the absolute value of the rate change; the rate threshold and the SNR threshold are set for the average frequency hopping rate and the signal-to-noise ratio mean value of all sub-frequency bands; the modulation modes of the PSK phase modulation are respectively represented as BPSK modulation, QPSK modulation and 8-PSK modulation according to 2 phases, 4 phases and 8 phases; When the frequency hopping average rate of the frequency hopping signal in the sub-frequency band is higher than the rate threshold and the signal-to-noise ratio average is lower than the SNR threshold, the PSK phase modulation uses BPSK modulation; When the average frequency hopping rate of the frequency hopping signal in the sub-frequency band is higher than the rate threshold and the average signal-to-noise ratio is higher than the SNR threshold, or when the average frequency hopping rate of the frequency hopping signal in the sub-frequency band is lower than the rate threshold and the average signal-to-noise ratio is lower than the SNR threshold, the PSK phase modulation uses QPSK modulation; When the average frequency hopping rate of the frequency hopping signal in the sub-frequency band is lower than the rate threshold and the average signal-to-noise ratio is higher than the SNR threshold, the PSK phase modulation uses 8-PSK modulation.

[0008] Furthermore, the process of setting the rate threshold includes: setting a number of rate sequence points along the time axis of the frequency change curve in each sub-frequency band, and marking the rate value at each rate sequence point based on the frequency change curve; using the K-means clustering analysis method to divide all rate sequence points into k clusters and perform K-means clustering to generate a centroid in each cluster, the centroid represents the average value of the rate values ​​of all rate sequence points in the corresponding cluster, and the mean of the two centroids with the largest interval is set as the rate threshold.

[0009] Furthermore, based on the elbow rule, the k value is corrected for the optimal k value, and the process includes: marking the k value used in the initial execution of K-means clustering as the initial k value, and marking the generated centroid as the initial centroid; selecting a number of values ​​on both sides of the numerical axis around the initial k value as k value elements, calculating the sum of squared errors between the rate values ​​of the rate sequence points corresponding to each k value element and the initial centroid, marking all the calculated sum of squared errors on a plane coordinate system with the abscissa being the k value, and connecting discrete sum of squared error value points into a curve, setting the k value corresponding to the sum of squared errors closest to the inflection point of the curve as the corrected optimal k value, and using the optimal k value in the K-means clustering analysis method to regenerate the centroid for calculating the rate threshold.

[0010] Furthermore, the calculation process of the sum of squared errors is as follows: the sum of squared errors is represented as SSE, the rate sequence point is represented as x, the total number and ordinal number of the rate sequence points are represented as n and i respectively, and the centroid is represented as μ, The calculation formula of the error sum of squares is expressed as: , The k value is selected from the k value element, x i represents the i-th rate sequence point, μ k Represents the center of mass at the current k value.

[0011] Furthermore, the phase-locked loop frequency tracking module includes: A phase comparator, used to calculate the phase difference between the input signal and the local oscillation signal; A low-pass filter is used to filter the high-frequency noise output by the phase comparator to generate a smooth control signal; A local oscillator receives a control voltage from the phase comparator output to adjust an output frequency of an input signal; Wherein, the phase comparator, the low-pass filter and the local oscillator are signal-connected in sequence.

[0012] Furthermore, the phase comparator is an edge detection type phase comparator, which is used to detect the zero crossing position of the input signal and the VCO output signal, and calculate the time difference between the two to obtain the phase error value.

[0013] Furthermore, the process of acquiring the zero-crossing point position is set to use the least square method to perform interpolation fitting between three consecutive sampling points to obtain a quadratic polynomial, and the zero-crossing point position is calculated based on the quadratic polynomial.

[0014] Further, the local oscillator is provided with a VCO reference frequency for dynamically adjusting the output frequency of the modulation signal; When the output frequency of the modulation signal is lower than the VCO reference frequency, the phase comparator generates a positive control signal; When the output frequency of the modulation signal is higher than the VCO reference frequency, the phase comparator generates a negative control signal; When the output frequency of the modulation signal is equal to the VCO reference frequency, the phase comparator generates a zero-value control signal.

[0015] Furthermore, an upper boundary value and a lower boundary value of each sub-frequency band are both provided with an offset redundancy.

[0016] Compared with the prior art, the present invention transfers the target frequency hopping signal to the frequency domain for modulation optimization through PSK phase keying modulation, and uses the phase-locked loop technology to frequency synchronize the modulation signal of each sub-band, so that the phase-locked loop can dynamically adjust and correct the VCO voltage-controlled oscillator frequency, ensuring the frequency consistency of each sub-band. It has the advantages of avoiding phase drift caused by frequency mismatch and obtaining the beneficial effect of higher-precision signal detection in the target filter. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the embodiments of the present invention, constitute a part of this application, and do not constitute a limitation of the embodiments of the present invention. In the drawings: Figure 1 It is a flowchart of the present invention. DETAILED DESCRIPTION

[0018] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with embodiments and drawings. The exemplary embodiments of the present invention and their description are only used to explain the present invention and are not intended to limit the present invention. Example

[0019] like Figure 1 As shown, this embodiment is a frequency hopping signal detection method for a filter, the method comprising: Step S1: Select a target filter and a target frequency hopping signal for frequency hopping, collect historical frequency hopping data of the target frequency hopping signal and use it to set an expected frequency hopping range, and divide the expected frequency hopping range of the target frequency hopping signal into a plurality of sub-frequency bands; Step S2: starting the frequency hopping process of the target filter, using PSK phase modulation to convert the target frequency hopping signal to the frequency domain and then performing phase modulation along each sub-band to generate a modulated signal, and maintaining smooth transition phase control between adjacent phase modulations; Step S3: construct a phase-locked loop frequency tracking module, and use the phase-locked loop frequency tracking module to perform VCO frequency modulation on the modulation signal of each sub-band to complete the frequency synchronization of the modulation signal in the phase modulation process of each sub-band; Step S4: convert all sub-bands output after VCO frequency modulation into time domain through PSK phase demodulation and mark them as modified frequency hopping signals, apply the modified frequency hopping signals to the target filter and re-implement the detection of the frequency hopping signals.

[0020] The target filter represents a filter device that is currently being greatly affected by the time response delay, resulting in a decrease in the system capture rate, which causes more phase errors in the signal, and thus significantly reduces the accuracy of signal detection during the frequency hopping process. The target frequency hopping signal refers to the signal that is to be frequency-hopped in the system. A frequency hopping signal is usually a signal containing multiple frequency points, and its frequency is changed by "hopping" to avoid interference, improve confidentiality or improve signal reliability. In practical applications, the frequency hopping signal may be a carrier signal in wireless communication, or a signal used in the fields of radar, navigation, radio spectrum detection, etc. "Signal frequency hopping" refers to a signal that jumps according to specific rules within a predetermined frequency range. For example, in a frequency hopping communication system, the carrier signal will switch quickly between multiple frequency points. This frequency hopping operation is usually to prevent spectrum congestion, improve anti-interference ability or increase signal security. Here, the target frequency hopping signal for frequency hopping refers to a signal that undergoes frequency changes under filter processing.

[0021] The expected frequency hopping range means the frequency range of the target frequency hopping signal in the frequency hopping process, which is set in advance according to the application requirements and historical data of the target system. The expected frequency hopping range includes the lowest frequency and the highest frequency of the frequency hopping, and is divided into multiple sub-frequency bands. The frequency range of each sub-frequency band can be regarded as an independent working interval in the frequency hopping process, and the goal is to ensure that the frequency changes smoothly and the phase error is minimized within each sub-frequency band. The PSK phase modulation is a digital modulation technology that represents different digital information by changing the phase of the signal carrier; PSK modulation maps digital data to different phases of the carrier by discretely changing the phase of the carrier, thereby realizing data transmission. Using a phase-locked loop for VCO frequency modulation means controlling and adjusting the VCO frequency through a phase-locked loop PLL to ensure frequency synchronization of the frequency hopping signal between different frequency bands; the role of the phase-locked loop is to provide stable and accurate frequency tracking during the frequency hopping process, and to achieve frequency synchronization to reduce the phase error caused by frequency inconsistency.

[0022] The frequency hopping signal is started and the frequency hopping process begins. The frequency hopping signal is transferred to the frequency domain using PSK phase modulation, and phase modulation is performed on each sub-band. The purpose is to map the phase information of the original signal to the frequency domain. The frequency hopping signal in each sub-band will be modulated into a signal with a specific phase. It is necessary to smoothly control the phase transition between adjacent frequency bands to avoid mutations and phase jumps caused by frequency band switching. In the common technology in this field, exponential smooth transition or sinusoidal transition function can be used for optimization. The phase-locked loop frequency tracking module is used to synchronize the frequency of the modulation signal of each sub-band and modulate the VCO frequency to maintain frequency consistency. The phase-locked loop ensures that the frequency of each sub-band is consistent with the reference signal of the phase-locked loop by adjusting the frequency of the VCO in real time. The working principle of the PLL is to adjust the output frequency of the VCO by comparing the phase of the input signal with the phase of the local reference signal so that it is synchronized with the frequency of the input signal. Then, the modulated signal is demodulated by PSK phase demodulation, and the signal is converted from the frequency domain back to the time domain to obtain a corrected frequency hopping signal. The previous PSK modulation and PLL synchronization process has corrected the phase error, and the demodulated frequency hopping signal will have less phase error than the original signal to achieve the beneficial effect of obtaining higher-precision signal detection in the target filter.

[0023] As a feasible implementation method, the upper boundary value and the lower boundary value of each sub-frequency band are provided with an offset redundancy. The offset redundancy refers to an additional frequency range added to the upper boundary value and the lower boundary value of each sub-frequency band. The redundancy is a dynamic and adjustable value that can help compensate for possible frequency errors, drifts or system response delays. The system may be affected by factors such as noise, frequency errors, and equipment accuracy. The offset redundancy provides a buffer to tolerate these uncertainties. During high-speed frequency hopping or multi-band hopping, the frequency of the signal may fluctuate or exceed the predetermined boundaries in a small range due to various reasons. Adding offset redundancy can ensure that the frequency band can still be effectively detected and processed within these fluctuation ranges.

[0024] Further, as a feasible implementation method, the process of modulating each sub-frequency band using PSK phase modulation includes: the historical frequency hopping data includes a signal-to-noise ratio mean value and a frequency change curve corresponding to the xy coordinate axis representing time-frequency, the total amount of the absolute value of the frequency change is collected for each sub-frequency band in the frequency change curve, and the total frequency hopping duration is obtained based on the time axis on the horizontal axis of the frequency change curve; the average frequency hopping rate of the target frequency hopping signal is calculated in each sub-frequency band according to the total frequency hopping duration and the total amount of the absolute value of the rate change; the rate threshold and the SNR threshold are set for the average frequency hopping rate and the signal-to-noise ratio mean value of all sub-frequency bands; the modulation modes of the PSK phase modulation are respectively represented as BPSK modulation, QPSK modulation and 8-PSK modulation according to 2 phases, 4 phases and 8 phases; When the frequency hopping average rate of the frequency hopping signal in the sub-frequency band is higher than the rate threshold and the signal-to-noise ratio average is lower than the SNR threshold, the PSK phase modulation uses BPSK modulation; When the average frequency hopping rate of the frequency hopping signal in the sub-frequency band is higher than the rate threshold and the average signal-to-noise ratio is higher than the SNR threshold, or when the average frequency hopping rate of the frequency hopping signal in the sub-frequency band is lower than the rate threshold and the average signal-to-noise ratio is lower than the SNR threshold, the PSK phase modulation uses QPSK modulation; When the average frequency hopping rate of the frequency hopping signal in the sub-frequency band is lower than the rate threshold and the average signal-to-noise ratio is higher than the SNR threshold, the PSK phase modulation uses 8-PSK modulation.

[0025] The mean signal-to-noise ratio refers to the average value of the ratio of signal to noise intensity during the historical operation period. In the frequency hopping system, the signal-to-noise ratio is a key indicator for measuring signal quality, which reflects the strength of the received signal relative to the background noise. The mean signal-to-noise ratio can be obtained by averaging the SNR values ​​within a certain time period. For example, for the signal in each frequency band or time window, the signal-to-noise ratio at each moment can be calculated, and then these values ​​are averaged to obtain the mean signal-to-noise ratio of the time period or frequency band. The frequency change curve is used to represent the trajectory of the frequency change of the frequency hopping signal over time in the time-frequency two-dimensional coordinate system, with the horizontal axis being time and the vertical axis being frequency; the curve shows the change of the signal frequency over time during the frequency hopping process. The total amount of the absolute value of the frequency change refers to the total change in the amplitude of the signal frequency change in each corresponding sub-frequency band; the absolute value of the frequency change is a scalar, which only means taking the absolute value of the "amplitude difference between two adjacent frequency values", regardless of positive and negative cancellation, and the meaning of the total amount is the cumulative change of the signal frequency. For example, in a jump, the frequency jumps from positive 50KHz to negative 50KHz and then to positive 20KHz. The absolute value of the frequency change between positive 50KHz and negative 50KHz is |50-(-50)|=100KHz, and the absolute value of the frequency change between negative 50KHz and positive 20KHz is |-50-20|=70KHz. The total absolute value of the frequency change in the enumeration process is 100+70=170KHz. In the frequency change curve, the total frequency hopping duration of the target filter is obtained by the horizontal coordinate in the time-frequency two-dimensional coordinate system. Then, by combining the total absolute value of the frequency change with the total frequency hopping duration, the average frequency hopping rate of the frequency hopping signal is calculated = the total absolute value of the frequency change / the total frequency hopping duration. The average frequency hopping rate reflects the speed of frequency change in a certain sub-band.

[0026] When the frequency hopping rate is high, that is, the frequency of the frequency hopping signal changes rapidly, it may cause a large phase change, thereby increasing the demodulation difficulty of the system. When the average frequency hopping rate of the frequency hopping signal in the sub-band is high (that is, the frequency of the signal changes very rapidly) and the mean signal-to-noise ratio is low (that is, the signal quality is poor and the noise is strong), the system selects BPSK modulation to ensure that the signal can still be transmitted stably under the conditions of high-speed frequency hopping and low signal-to-noise ratio, avoiding excessive phase errors and bit errors; fast frequency hopping at high frequency hopping rates may cause large phase errors, and BPSK modulation, as the most basic and noise-resistant modulation method, can provide better fault tolerance under such conditions; under low signal-to-noise ratio conditions, higher-order modulation modes are easily affected by noise, resulting in higher bit error rates, so BPSK has only two possible phases of 0° and 180°, which makes the phase error have less impact on the demodulation process, and its lower order enables it to better adapt to low signal-to-noise ratio environments, reduce the risk of bit errors, and ensure reliable signal reception.

[0027] When the average frequency hopping rate is higher than the rate threshold and the mean signal-to-noise ratio is higher than the SNR threshold, it means that the frequency hopping signal switches rapidly in the current sub-frequency band, the frequency hopping rate is high, the signal instability is large, and the frequency change is fast, which easily introduces phase error and interference, but the signal quality is good, and the signal is strong and stable; low average rate means good signal stability, because the frequency changes slowly, the signal will not change rapidly, so it is relatively easy to process. When the average frequency hopping rate is lower than the rate threshold and the mean signal-to-noise ratio is lower than the SNR threshold, it means that the frequency change of the frequency hopping signal in the sub-frequency band is relatively slow, the frequency switching speed is low, because the frequency changes slowly, the signal will not change rapidly and the stability is good, and at the same time, due to the low mean signal-to-noise ratio, it indicates that the signal quality is poor, the background noise is strong, and the signal may be greatly interfered at the receiving end. Therefore, the signal is easily affected by noise during transmission, which may result in a high bit error rate. Based on this, when the frequency hopping rate is low and the signal-to-noise ratio is high, or the frequency hopping rate is high but the signal-to-noise ratio is low, this implementation method selects QPSK modulation. Compared with BPSK modulation, the QPSK modulation can provide a higher data transmission rate and is more stable than higher phase modulation modes. When the frequency hopping rate is high and the signal-to-noise ratio is good, QPSK modulation can provide a balanced transmission rate and anti-interference capability; on the other hand, even if the frequency hopping rate is low and the signal-to-noise ratio is poor, QPSK modulation can still provide higher transmission efficiency and anti-interference capability. Compared with BPSK or higher phase modulation, QPSK still has better performance under lower signal-to-noise ratio conditions. The QPSK modulation scheme under this condition aims to provide a compromise solution, that is, when the frequency hopping is high and the signal quality is good, the high transmission efficiency of QPSK is fully utilized; when the frequency hopping is low and the signal quality is poor, the selection of QPSK can still provide good anti-interference performance while ensuring the improvement of transmission efficiency.

[0028] When the average frequency hopping rate of the frequency hopping signal in the sub-band is lower than the rate threshold and the average signal-to-noise ratio is higher than the SNR threshold, the phase change of the signal is relatively stable and is not easily affected by frequent frequency hopping. At the same time, a high signal-to-noise ratio indicates good signal quality and low noise interference. The system can effectively identify more phase changes, so it is suitable for using higher-order modulation methods. 8-PSK modulation uses 8 different phases to represent data. Compared with BPSK's 2 phases and QPSK's 4 phases, 8-PSK can transmit more data under the same bandwidth conditions, which can improve the efficiency and capacity of data transmission while ensuring signal stability.

[0029] Furthermore, as a feasible implementation method, the process of setting the rate threshold includes: setting a number of rate sequence points along the time axis of the frequency change curve in each sub-frequency band, and marking the rate value at each rate sequence point based on the frequency change curve; using the K-means clustering analysis method to divide all rate sequence points into k clusters and perform K-means clustering to generate a centroid in each cluster, the centroid represents the average value of the rate values ​​of all rate sequence points in the corresponding cluster, and the mean of the two centroids with the largest interval is set as the rate threshold.

[0030] On the time axis of the frequency change curve, several rate sequence points are selected. These rate sequence points represent the frequency hopping rate at different time moments, that is, each rate sequence point corresponds to the speed of the signal frequency change at a specific time point. K-means cluster analysis is a common unsupervised learning algorithm used to divide a data set into several clusters. In this embodiment, the goal of K-means cluster analysis is to divide all rate sequence points according to the similarity of their rate values ​​and generate multiple clusters. The centroid represents the average rate of the rate value in each cluster. The two centroids with the largest interval are selected because these two centroids represent the two extreme values ​​of the frequency hopping rate, namely high-speed frequency hopping and low-speed frequency hopping. These two extreme values ​​can effectively define different intervals of the frequency hopping rate. The centroid with the largest interval can fully reflect the range of change in the rate of the frequency hopping signal. When setting the rate threshold, using the mean of these two centroids helps to find a suitable dividing point between high-speed and low-speed frequency hopping.

[0031] Furthermore, as a feasible implementation method, the k value is corrected to the optimal k value based on the elbow rule, and the process includes: marking the k value used in the initial execution of K-means clustering as the initial k value, and marking the generated centroid as the initial centroid; selecting a number of values ​​on both sides of the numerical axis of the initial k value as k value elements, calculating the sum of squared errors between the rate values ​​of the rate sequence points corresponding to each k value element and the initial centroid, marking all the calculated sum of squared errors on a plane coordinate system with the horizontal axis being the k value, and connecting the discrete sum of squared error value points into a curve, setting the k value corresponding to the sum of squared error value closest to the inflection point of the curve as the corrected optimal k value, and using the optimal k value in the K-means clustering analysis method to regenerate the centroid for calculating the rate threshold.

[0032] Use a certain initial k value to perform K-means clustering to obtain the clustering results and the corresponding centroid. The initial k value can be estimated based on domain knowledge or previous experience to obtain a suitable initial k value. The k value element refers to selecting multiple different k values ​​(i.e., the number of clusters) within a certain range as the number of candidate clusters based on the initial k value. The purpose is to explore the changes in clustering effect and error sum of squares under different numbers of clusters. For a series of different k values ​​in the k value element, calculate the error sum of squares corresponding to each k value. These error sum of squares will form a set of discrete numerical points; use these error sum of squares and the corresponding k values ​​as coordinates to draw a curve in the plane coordinate system. The horizontal axis represents the number of clusters k, and the vertical axis represents the error sum of squares. Each k value corresponds to a error sum of squares, and the discrete numerical points will be connected into a curve. As the k value increases, the error sum of squares will gradually decrease, showing a decreasing trend; but as the k value increases to a certain extent, the decrease in the error sum of squares will slow down significantly, and the curve will form an inflection point, that is, the elbow position in the curve. By observing the curve, we can find the location of the elbow, where the rate of decrease of the error sum of squares slows down significantly. The elbow usually represents the best point for clustering, because after this point, adding more clusters will not significantly improve the clustering effect. According to the elbow rule, the k value at the inflection point of the curve is selected as the optimal number of clusters. Specifically, by measuring the distance between each point in the curve and the elbow position, the error sum of squares corresponding to the k value closest to the inflection point can be selected.

[0033] Further, as a feasible implementation, the calculation process of the sum of squared errors is as follows: the sum of squared errors is represented as SSE, the rate sequence point is represented as x, the total number and ordinal number of the rate sequence points are represented as n and i respectively, and the centroid is represented as μ, The calculation formula of the error sum of squares is expressed as: , The k value is selected from the k value element, x i represents the i-th rate sequence point, μ k Represents the center of mass at the current k value.

[0034] The error sum square value SSE is an important indicator in cluster analysis, which is used to measure the average deviation of the data points in the cluster to the cluster centroid; the smaller the value, the closer the data point is to the cluster centroid, and the better the clustering effect. In this embodiment, the error sum square value SSE represents the average deviation of all rate sequence points in the kth cluster from their cluster centroid μ k The sum of the squared differences between represents the squared difference between each rate sequence point and its cluster centroid, and all these squared differences are added together to obtain the sum of squared errors for the cluster. krepresents the centroid of the kth cluster, which is the average of all rate sequence points in the cluster. The centroid is determined by calculating the average position of all data points in the cluster. k The calculation formula can be expressed as: , where n k Represents the number of rate sequence points of the kth cluster.

[0035] Furthermore, the phase-locked loop frequency tracking module includes: A phase comparator, used to calculate the phase difference between the input signal and the local oscillation signal; A low-pass filter is used to filter the high-frequency noise output by the phase comparator to generate a smooth control signal; A local oscillator receives a control voltage from the phase comparator output to adjust an output frequency of an input signal; Wherein, the phase comparator, the low-pass filter and the local oscillator are signal-connected in sequence.

[0036] The phase comparator is used to calculate the phase difference between the input signal and the local oscillator signal. It compares the input modulated signal with the local reference oscillator signal and outputs a voltage signal proportional to their phase difference. If the two signals match in phase, the phase comparator output is zero. If a phase difference exists, the output voltage will increase as the phase difference increases. The output of the phase comparator is a changing signal that represents the phase change of the input signal relative to the reference signal. The main task of the low-pass filter is to remove high-frequency noise from the output of the phase comparator and generate a smooth control signal; the output signal of the phase comparator usually contains some high-frequency components or noise, and the low-pass filter can filter out these high-frequency components and only retain the low-frequency part, thereby ensuring that the output control signal is smooth. By smoothing the output, the low-pass filter helps stabilize the frequency control, making the control signal more suitable for adjusting the local oscillator. This smooth control signal can directly affect the frequency adjustment of the oscillator. The local oscillator receives the smoothed control voltage output by the low-pass filter and adjusts its output frequency according to this control voltage. The local oscillator is the core part of the frequency adjustment in the system. It tracks the frequency changes of the input signal by changing its output frequency. When the control signal changes, the frequency of the local oscillator is also adjusted accordingly, thereby ensuring that the frequency of the input signal is synchronized with the local oscillation signal. The frequency of the local oscillator is adjusted by the control voltage provided by the low-pass filter. The change of the control voltage directly affects the operating frequency of the oscillator. If the frequency of the input signal changes, the low-pass filter will adjust the control signal, thereby changing the output frequency of the local oscillator, and finally achieving frequency synchronization between the input signal and the local oscillation signal.

[0037] Furthermore, as a feasible implementation method, the phase comparator adopts an edge detection type phase comparator, which is used to detect the zero crossing position of the input signal and the VCO output signal, and obtain the phase error value after calculating the time difference between the two; the process of obtaining the zero crossing position is set to use the least squares method to perform interpolation fitting between three consecutive sampling points to obtain a quadratic polynomial, and the zero crossing position is calculated based on the quadratic polynomial.

[0038] The edge detection phase comparator is used to detect the zero crossing of the signal, that is, the moment when the signal changes from positive to negative, or from negative to positive. The detection of the zero crossing is the basis for determining the phase difference of the signal, because the position of the zero crossing is directly related to the periodic change and phase characteristics of the signal. The input signal and the VCO output signal detect their zero crossings through the phase comparator, and the time difference of the zero crossing is calculated; the time difference of the zero crossing is the phase difference between the input signal and the VCO output signal, and this phase difference can be calculated by the relationship between the time difference and the signal period. In order to improve the accuracy of the zero crossing position, the edge detection phase comparator uses the least squares method to interpolate and fit between three consecutive sampling points. The least squares method is an optimization algorithm that aims to minimize the sum of squared errors and find the most suitable quadratic polynomial to approximate the change trend of these three points. The three sampling points refer to three consecutive data points of the input signal or the VCO output signal within the sampling period. Based on these three sampling points, a quadratic polynomial is fitted using the least squares method, which can accurately represent the change of the signal between these three points. In specific applications, the quadratic polynomial can be set to f(t)=λt 2 +βt+φ, where λ, β and φ are fitting parameters obtained by the least squares method, describing the trend of the signal between these three points. Using the obtained quadratic polynomial, its zero crossing position can be solved. The zero crossing position refers to the solution of the polynomial, which can be set as: λt 2 +βt+φ=0. By solving the above equation, the time t0 of the zero crossing point is obtained, that is, the zero crossing time of the input signal and the VCO output signal. By fitting a quadratic polynomial between three sampling points using the least squares method, a more accurate zero crossing position can be obtained than a single sampling point. This method can reduce the error caused by uneven sampling points or noise and improve the accuracy of phase comparison. Interpolation fitting between consecutive sampling points can effectively reduce the influence of sampling noise on the zero crossing position and avoid large phase errors caused by single sampling point errors.

[0039] Further, as a feasible implementation, the local oscillator is provided with a VCO reference frequency for dynamically adjusting the output frequency of the modulation signal; When the output frequency of the modulation signal is lower than the VCO reference frequency, the phase comparator generates a positive control signal; When the output frequency of the modulation signal is higher than the VCO reference frequency, the phase comparator generates a negative control signal; When the output frequency of the modulation signal is equal to the VCO reference frequency, the phase comparator generates a zero-value control signal.

[0040] The VCO reference frequency is the reference frequency of the local oscillator, which is used to set the target frequency. When the frequency of the modulated signal needs to be synchronized with the reference frequency, the local oscillator adjusts its output frequency according to the feedback of the phase comparator. The output frequency of the modulated signal is the frequency output by the local oscillator, which is usually synchronized or adjusted with the input signal. The frequency change of the modulated signal is controlled by the phase-locked loop and adjusted according to the output of the phase comparator to achieve frequency synchronization. When the output frequency of the modulated signal is lower than the VCO reference frequency, it indicates that the frequency of the modulated signal is lower and the output frequency needs to be increased to match the reference frequency. The phase comparator will generate a positive control signal, which will be fed back to the local oscillator to prompt it to increase the output frequency. When the output frequency of the modulated signal is higher than the VCO reference frequency, the output frequency needs to be reduced to achieve synchronization. The phase comparator generates a negative control signal, which will be fed back to the local oscillator to prompt it to reduce the output frequency. When the output frequency of the modulated signal is equal to the VCO reference frequency, the frequency of the modulated signal is completely consistent with the reference frequency and no adjustment is required. The phase comparator generates a zero-value control signal, which means that the output frequency of the VCO has completely matched the target frequency.

[0041] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A frequency hopping signal detection method for a filter, characterized in that: The method includes: Step S1: Select a target filter and a target frequency hopping signal for frequency hopping, collect historical frequency hopping data of the target frequency hopping signal and use it to set an expected frequency hopping range, and divide the expected frequency hopping range of the target frequency hopping signal into a plurality of sub-frequency bands; Step S2: starting the frequency hopping process of the target filter, using PSK phase modulation to convert the target frequency hopping signal to the frequency domain and then performing phase modulation along each sub-band to generate a modulated signal, and maintaining smooth transition phase control between adjacent phase modulations; Step S3: construct a phase-locked loop frequency tracking module, and use the phase-locked loop frequency tracking module to perform VCO frequency modulation on the modulation signal of each sub-band to complete the frequency synchronization of the modulation signal in the phase modulation process of each sub-band; Step S4: convert all sub-bands output after VCO frequency modulation into time domain through PSK phase demodulation and mark them as modified frequency hopping signals, apply the modified frequency hopping signals to the target filter and re-implement the detection of the frequency hopping signals.

2. A frequency hopping signal detection method for a filter according to claim 1, characterized in that: The process of modulating each sub-frequency band using PSK phase modulation includes: the historical frequency hopping data includes a signal-to-noise ratio mean value and a frequency change curve corresponding to the xy coordinate axis representing time-frequency, the total amount of the absolute value of the frequency change is collected for each sub-frequency band in the frequency change curve, and the total frequency hopping duration is obtained based on the time axis on the abscissa of the frequency change curve; the average frequency hopping rate of the target frequency hopping signal is calculated in each sub-frequency band according to the total frequency hopping duration and the total amount of the absolute value of the rate change; the rate threshold and the SNR threshold are set for the average frequency hopping rate and the signal-to-noise ratio mean value of all sub-frequency bands; the modulation modes of the PSK phase modulation are respectively represented as BPSK modulation, QPSK modulation and 8-PSK modulation according to 2 phases, 4 phases and 8 phases; When the frequency hopping average rate of the frequency hopping signal in the sub-frequency band is higher than the rate threshold and the signal-to-noise ratio average is lower than the SNR threshold, the PSK phase modulation uses BPSK modulation; When the average frequency hopping rate of the frequency hopping signal in the sub-frequency band is higher than the rate threshold and the average signal-to-noise ratio is higher than the SNR threshold, or when the average frequency hopping rate of the frequency hopping signal in the sub-frequency band is lower than the rate threshold and the average signal-to-noise ratio is lower than the SNR threshold, the PSK phase modulation uses QPSK modulation; When the average frequency hopping rate of the frequency hopping signal in the sub-frequency band is lower than the rate threshold and the average signal-to-noise ratio is higher than the SNR threshold, the PSK phase modulation uses 8-PSK modulation.

3. A frequency hopping signal detection method for a filter according to claim 2, characterized in that: The process of setting the rate threshold includes: setting a number of rate sequence points along the time axis of the frequency change curve in each sub-frequency band, marking the rate value at each rate sequence point based on the frequency change curve; using the K-means clustering analysis method to divide all rate sequence points into k clusters and perform K-means clustering to generate a centroid in each cluster, the centroid represents the average value of the rate values ​​of all rate sequence points in the corresponding cluster, and setting the mean of the two centroids with the largest interval as the rate threshold.

4. A frequency hopping signal detection method for a filter according to claim 3, characterized in that: The k value is corrected for the optimal k value based on the elbow rule, and the process includes: marking the k value used in the initial execution of K-means clustering as the initial k value, and marking the generated centroid as the initial centroid; selecting a number of values ​​on both sides of the number axis around the initial k value as k value elements, calculating the sum of squared errors between the rate values ​​of the rate sequence points corresponding to each k value element and the initial centroid, marking all the calculated sum of squared errors on a plane coordinate system with the abscissa being the k value, and connecting discrete sum of squared error value points into a curve, setting the k value corresponding to the sum of squared errors closest to the inflection point of the curve as the corrected optimal k value, and using the optimal k value in a K-means clustering analysis method to regenerate the centroid for calculating a rate threshold.

5. A frequency hopping signal detection method for a filter according to claim 4, characterized in that: The calculation process of the square sum of errors is as follows: the square sum of errors is represented as SSE, the rate sequence point is represented as x, the total number and ordinal number of the rate sequence points are represented as n and i respectively, and the centroid is represented as μ. The calculation formula of the error sum of squares is expressed as: , The k value is selected from the k value element, x i represents the i-th rate sequence point, μ k Represents the center of mass at the current k value.

6. A frequency hopping signal detection method for a filter according to claim 1, characterized in that: The phase-locked loop frequency tracking module comprises: A phase comparator, used to calculate the phase difference between the input signal and the local oscillation signal; A low-pass filter is used to filter the high-frequency noise output by the phase comparator to generate a smooth control signal; A local oscillator receives a control voltage from the phase comparator output to adjust an output frequency of an input signal; Wherein, the phase comparator, the low-pass filter and the local oscillator are signal-connected in sequence.

7. A frequency hopping signal detection method for a filter according to claim 6, characterized in that: The phase comparator is an edge detection type phase comparator, which is used to detect the zero crossing position of the input signal and the VCO output signal, and calculate the time difference between the two to obtain the phase error value.

8. A frequency hopping signal detection method for a filter according to claim 7, characterized in that: The process of acquiring the zero-crossing point position is as follows: using the least square method to perform interpolation fitting between three consecutive sampling points to obtain a quadratic polynomial, and calculating and acquiring the zero-crossing point position based on the quadratic polynomial.

9. A frequency hopping signal detection method for a filter according to claim 6, characterized in that: The local oscillator is provided with a VCO reference frequency for dynamically adjusting the output frequency of the modulation signal; When the output frequency of the modulation signal is lower than the VCO reference frequency, the phase comparator generates a positive control signal; When the output frequency of the modulation signal is higher than the VCO reference frequency, the phase comparator generates a negative control signal; When the output frequency of the modulation signal is equal to the VCO reference frequency, the phase comparator generates a zero-value control signal.

10. A frequency hopping signal detection method for a filter according to any one of claims 1 to 9, characterized in that: An upper boundary value and a lower boundary value of each sub-frequency band are both provided with an offset redundancy.

Images