Radiation source signal detection method and system

Abstract

The invention provides a radiation source signal detection method, which comprises the following steps of: S1, carrying out noise floor estimation on a received radiation source signal, and de-noising; s2, estimating the center frequency and the bandwidth of the denoised signal; and S3, extracting a center frequency and a bandwidth part in the signal as a detection signal. According to the technology of the invention, the center frequency and bandwidth of a plurality of signals are accurately estimated by using a peak searching algorithm, and spurious signals can be effectively filtered out.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a radiation source signal detection method and system. Background Art

[0002] With the development of information technology, the number and types of electromagnetic signals propagating in space are increasing. On the other hand, the instantaneous receiving bandwidth of the receiver is also increasing, resulting in multiple different signals entering the broadband receiver at the same time, and the power, bandwidth, modulation mode and other parameters of these signals are different. When further analyzing the time, space and frequency domain parameters of these signals, it is necessary to separate the different signals. The basic method of multi-signal separation is to use a suitable bandpass filter for filtering, which involves two key steps: one is signal detection, that is, detecting and distinguishing multiple signals in the frequency domain; the other is parameter estimation, estimating the center frequency and bandwidth of each signal, and designing a suitable bandpass filter based on these two parameters.

[0003] Generally speaking, signal detection can be divided into cooperative signal detection and non-cooperative signal detection. In electronic reconnaissance systems, detection is mainly aimed at non-cooperative signals. The difficulty of non-cooperative detection lies in the fact that the frequency band, occurrence time, azimuth, signal modulation style, channel characteristics and noise characteristics of the signal are all unknown, and its detection has a blind source characteristic. In electronic reconnaissance systems, due to the current complex electromagnetic environment, the detection of signals is usually the simultaneous detection of multiple signals. In practice, due to the unknown number of signals, the missed detection of low signal-to-noise ratio signals and the false detection of noise glitches in multi-signal detection will be more serious. In addition, the continuous development of informatization and intelligence also requires signal detection to have the characteristics of automation and efficiency. Based on these new demands and application backgrounds, the present invention aims to solve the problem of automated and efficient detection of multiple signals in non-cooperative electronic reconnaissance, and estimate the center frequency and bandwidth of each detected signal for subsequent more detailed analysis of each signal.

[0004] The Chinese patent document with the publication number CN116736680B discloses a method for removing solid tide response based on improved variational modal decomposition. It includes: inputting the data collected by the ground instrument and performing data preprocessing; calculating the spectrum of the data and comparing the frequency characteristics of the solid tide response to determine the influence range of the solid tide response on the data in the frequency domain; determining the frequency division point f according to the influence range of the solid tide response in the frequency domain; constraining the estimated center frequency of the variational modal decomposition according to the frequency division point f; using the variational modal decomposition method with center frequency constraint to decompose the data, find the optimal components under different center frequency constraints to obtain each mode, and arrange the decomposed modes from small to large according to their center frequencies; remove the solid tide response mode in each mode. It is difficult to solve the problem that the existing methods can effectively remove the periodic response generated by the solid tide in the frequency domain without destroying the frequency characteristics of different source signals in the observed data. In the detection process, this patent does not have a denoising process, nor does it update the threshold in real time.

[0005] Therefore, a high-precision radiation source signal detection method and system are needed. Summary of the invention

[0006] In view of the defects in the prior art, an object of the present invention is to provide a radiation source signal detection method and system.

[0007] A radiation source signal detection method provided by the present invention includes:

[0008] Step S1: Estimating the noise floor of the received radiation source signal and denoising it;

[0009] Step S2: estimating the center frequency and bandwidth of the denoised signal;

[0010] Step S3: extract the center frequency and bandwidth of the signal as the detection signal.

[0011] Preferably, step S1 comprises:

[0012] Step S1.1: The collected power spectrum is S, the power spectrum value threshold is λ, and the original threshold is set to λ ini , downward adjustment coefficient a, downward adjustment coefficient b;

[0013] Where k is the index value of the power spectrum, a<1, b>1; the kth noise floor estimation value is N(k), and S(k) is the kth power spectrum value;

[0014] Step S1.2: Determine the sign of (S(k)-N(k-1))×(S(k)-N(k-2)). If the sign is positive, execute step S1.3; if the sign is negative, execute step S1.4;

[0015] Step S1.3: Determine the sign of S(k)-N(k-1). If the sign is positive, update the threshold to λ. New =λ×a; if the sign is negative, update the threshold to λ New =λ×b; after updating the threshold, execute step S1.5;

[0016] Step S1.4: Let the threshold λ New =λ ini , then execute step S1.5;

[0017] Step S1.5: Compare the absolute value of S(k)-N(k-1) with the threshold λ New , if the threshold λ New If the threshold λ is large, let the correction value e = S(k)-N(k-1); New If it is small, let e = sgn(S(k)-N(k-1))λ New , where sgn(x) represents the sign of x;

[0018] Step S1.6: Update the noise floor N(k)=N(k-1)+e, update the index value to k+1, return to step S1.2, and execute step S1.7 after estimating the noise floor of the entire power spectrum;

[0019] Step S1.7: In the collected power spectrum, remove the noise floor to complete denoising.

[0020] Preferably, step S2 comprises:

[0021] Step S2.1: Set the definition rule of the mark value, the peak mark is f peak , when in the peak range, f peak =1; when in the non-peak range, f peak =0; the boundary mark is f edge , when at the boundary, f edge =1; when it is not on the boundary, f edge =0;

[0022] Step S2.2: Determine the minimum value p of the local power spectrum value min and the maximum value p max ; Specifically, when S(k)>p max And f peak =1, the local maximum value is updated to p max =S(k), otherwise p is not updated max ; When S(k) <p min And f peak = 0, update the local minimum value to p min =S(k), otherwise p is not updated min ;

[0023] Step S2.3: Estimate the start boundary and the end boundary of the signal according to the maximum and minimum values ​​of the local power spectrum values;

[0024] Step S2.4: Calculate the bandwidth and center frequency of the signal through the power spectrum value corresponding to the boundary, and set the peak mark f peak =1, boundary mark f edge =0;

[0025] Signal bandwidth = power spectrum value corresponding to the end boundary - power spectrum value corresponding to the start boundary

[0026] The center frequency of the signal = (the power spectrum value corresponding to the end boundary + the power spectrum value corresponding to the start boundary) / 2;

[0027] Step S2.5: Update the power spectrum index value to k+1. When the power index value is greater than the number of power spectrum values ​​M, proceed to step S3; otherwise return to execute steps S2.2-S2.4 until the power index value is greater than the number of power spectrum values ​​M.

[0028] Preferably, step S2.3 comprises:

[0029] Step S2.3.1: If S(k) <p max / 4 and f peak =1, let f peak =0,f edge =1, proceed to step S2.3.2; otherwise, proceed directly to step S2.3.3;

[0030] Step S2.3.2: If S(k)>4p min And f peak = 0, or the index value k is equal to the number of spectral values ​​M of the power spectrum and f edge =1, it indicates that the previous signal ends, and the current index value k is recorded as the current signal end boundary End1; otherwise, proceed to step S2.5;

[0031] Step S2.3.3: Update the signal boundary range; starting from the local maximum value, traverse the power spectrum value forward and backward, and when the power spectrum value of the signal is lower than the background noise for the first time, use the index value corresponding to the power spectrum value as the starting boundary Start and ending boundary End2 of the signal;

[0032] Step S2.3.4: Determine the final termination boundary End; compare the power spectrum values ​​corresponding to the termination boundary End1 and the termination boundary End2, take the termination boundary with the larger power spectrum value as the final termination boundary End, and set the termination boundary End as the starting boundary of the next signal.

[0033] Preferably, step S3 includes:

[0034] The central frequency and bandwidth parts of the radiation source signal are extracted, and the other parts of the signal are filtered out to obtain the detection signal.

[0035] A radiation source signal detection system provided according to the present invention comprises:

[0036] Module M1: estimates the noise floor of the received radiation source signal and performs denoising;

[0037] Module M2: Estimate the center frequency and bandwidth of the denoised signal;

[0038] Module M3: extract the center frequency and bandwidth of the signal as the detection signal.

[0039] Preferably, module M1 comprises:

[0040] Module M1.1: The collected power spectrum is S, the power spectrum value threshold is λ, and the original threshold is set to λ ini , downward adjustment coefficient a, downward adjustment coefficient b;

[0041] Where k is the index value of the power spectrum, a<1, b>1; the kth noise floor estimation value is N(k), and S(k) is the kth power spectrum value;

[0042] Module M1.2: Determine the sign of (S(k)-N(k-1))×(S(k)-N(k-2)). If the sign is positive, execute module M1.3; if the sign is negative, execute module M1.4.

[0043] Module M1.3: Determine the sign of S(k)-N(k-1). If the sign is positive, update the threshold to λ New =λ×a; if the sign is negative, update the threshold to λ New =λ×b; after updating the threshold, execute module M1.5;

[0044] Module M1.4: Let the threshold λ New =λ ini , then execute module M1.5;

[0045] Module M1.5: Compare the absolute value of S(k)-N(k-1) with the threshold λ New , if the threshold λ New If the threshold λ is large, let the correction value e = S(k)-N(k-1); New If it is small, let e = sgn(S(k)-N(k-1))λ New , where sgn(x) represents the sign of x;

[0046] Module M1.6: Update the noise floor N(k)=N(k-1)+e, update the index value to k+1, return to execute module M1.2, and execute module M1.7 after estimating the noise floor of the entire power spectrum;

[0047] Module M1.7: Remove the noise floor from the acquired power spectrum to complete denoising.

[0048] Preferably, module M2 comprises:

[0049] Module M2.1: Setting the definition rules of the marker value, the peak marker is f peak , when in the peak range, f peak =1; when in the non-peak range, f peak =0; the boundary mark is f edge , when at the boundary, f edge =1; when it is not on the boundary, f edge =0;

[0050] Module M2.2: Determine the minimum value p of the local power spectrum min and the maximum value p max ; Specifically, when S(k)>p max And f peak =1, the local maximum value is updated to p max =S(k), otherwise p is not updated max ; When S(k) <p min And f peak = 0, update the local minimum value to p min =S(k), otherwise p is not updated min ;

[0051] Module M2.3: Estimate the start and end boundaries of the signal based on the maximum and minimum values ​​of the local power spectrum;

[0052] Module M2.4: Calculate the bandwidth and center frequency of the signal through the power spectrum value corresponding to the boundary, and set the peak mark f peak =1, boundary mark f edge =0;

[0053] Signal bandwidth = power spectrum value corresponding to the end boundary - power spectrum value corresponding to the start boundary

[0054] The center frequency of the signal = (the power spectrum value corresponding to the end boundary + the power spectrum value corresponding to the start boundary) / 2;

[0055] Module M2.5: Update the power spectrum index value to k+1. When the power index value is greater than the number of power spectrum values ​​M, proceed to module M3; otherwise return to execute modules M2.2-M2.4 until the power index value is greater than the number of power spectrum values ​​M.

[0056] Preferably, module M2.3 includes:

[0057] Module M2.3.1: If S(k) <p max / 4 and f peak =1, let f peak =0,f edge =1, proceed to module M2.3.2; otherwise proceed directly to module M2.3.3;

[0058] Module M2.3.2: If S(k)>4p min And f peak = 0, or the index value k is equal to the number of spectral values ​​M of the power spectrum and f edge =1, it indicates that the previous signal ends, and the current index value k is recorded as the current signal end boundary End1; otherwise, module M2.5 is performed;

[0059] Module M2.3.3: Update the signal boundary range; starting from the local maximum value, traverse the power spectrum value forward and backward. When the power spectrum value of the signal is lower than the background noise for the first time, the index value corresponding to the power spectrum value is used as the starting boundary Start and ending boundary End2 of the signal;

[0060] Module M2.3.4: Determine the final termination boundary End; compare the power spectrum values ​​corresponding to the termination boundary End1 and the termination boundary End2, take the termination boundary with the larger power spectrum value as the final termination boundary End, and set the termination boundary End as the starting boundary of the next signal.

[0061] Preferably, module M3 comprises:

[0062] The central frequency and bandwidth parts of the radiation source signal are extracted, and the other parts of the signal are filtered out to obtain the detection signal.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] 1. The present invention uses the peak-finding algorithm to accurately estimate the center frequency and bandwidth of multiple signals, and can clearly detect the signal, and the burr signal is filtered out. In addition, it has a good detection and estimation effect on both large signal-to-noise ratio and small signal-to-noise ratio signals, and can effectively filter out stray signals;

[0065] 2. The present invention preferentially filters out noise by estimating the noise floor, and then continuously corrects the center frequency and bandwidth by the peak-finding method. Under the combined effect of the two, the accuracy and efficiency of signal detection are improved, and fully automatic detection is realized;

[0066] 3. The present invention combines the peak-finding algorithm with the noise floor estimation, which can effectively distinguish two adjacent signals when the signal spectra are close to each other. When faced with signal spectra such as nonlinear frequency modulation, it has a more accurate bandwidth estimation compared to a single peak-finding algorithm, thereby achieving accurate and effective detection of the signal. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0068] Figure 1 is a flow chart of the radiation source signal detection method of the present invention;

[0069] Figure 2 This is a flow chart of the signal center frequency and bandwidth estimation method in the present invention;

[0070] Figure 3 This is a schematic diagram of radiation source signal detection according to the present invention. DETAILED DESCRIPTION

[0071] The present invention is described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those of ordinary skill in the art, several changes and improvements can also be made without departing from the concept of the present invention. These all belong to the protection scope of the present invention.

[0072] The present invention considers non-cooperative multi-signal detection. First, a signal detection method based on power spectrum is adopted. The power spectrum is more stable than the FFT spectrum, has less fading, and is conducive to frequency domain signal detection. In terms of signal detection, there are mainly power spectrum single threshold detection method, double threshold detection method, spectral entropy method, etc.

[0073] Among them, the single threshold detection method is the fastest, but the detection performance deteriorates sharply under low signal-to-noise ratio. If the threshold is set too low, the false alarm rate will be too high, and if the threshold is set too high, there will be a large number of missed detections.

[0074] The double threshold detection method has good detection performance and fast detection speed in Gaussian white noise environment, but its performance deteriorates seriously in colored noise channel with fluctuating noise power spectrum.

[0075] The spectral entropy method has good noise resistance, but its detection speed cannot meet actual needs.

[0076] Recently, a newer detection method uses a combination of noise floor estimation and threshold detection to detect frequency domain signals. This method has a fast detection speed and adapts to fluctuating noise floors, but it does not perform well when faced with multiple signals with overlapping spectra.

[0077] The present invention adopts a peak-finding algorithm for the input frequency domain signal, regards the local minimum of the frequency domain signal as the start and end frequency points of the signal, and uses this to determine the possible start and end range of the signal. Subsequently, the start and end range of the signal is corrected in combination with the noise floor threshold detection method, and finally a relatively accurate signal range is obtained. The signal detection algorithm effectively addresses the problem of accurate detection of multiple signals through the peak-finding algorithm, and can effectively estimate the signal bandwidth, and has achieved remarkable results in practical applications.

[0078] The present invention still adopts a method of combining noise floor estimation and peak detection to perform signal detection and center frequency and bandwidth estimation.

[0079] Example 1

[0080] The noise floor estimation adopts the nonlinear autoregressive linear filtering method, and its input is the power spectrum S to be estimated. The specific process is as follows:

[0081] Step 1: Set the original threshold λ ini , initialize the threshold λ = λ ini , down-adjustment coefficient a, down-adjustment coefficient b, power spectrum index value k, noise floor estimation value N(1) = N(2) = min(S). (Note: Here, the initial N(1) and N(2) can also be a certain quantile value of the power spectrum S)

[0082] Step 2: Determine the sign of (S(k)-N(k-1))×(S(k)-N(k-2)). If the sign is positive, proceed to step 3); otherwise, proceed to step 4.

[0083] Step 3. Determine the sign of S(k)-N(k-1). If the sign is positive, update the threshold to λ=λa. This indicates that the frequency component values ​​of the power spectrum near this frequency band are on an upward trend. It can be inferred that this frequency band is the starting end of the signal power spectrum transition band. At this time, the step amplitude of the noise floor estimation value should be reduced to prevent over-estimation of the noise floor.

[0084] Otherwise, the updated threshold is λ=λb. Go to step 5. This indicates that the frequency components of the power spectrum decrease near this frequency band. It can be inferred that this is the end of the transition band of the signal power spectrum. A larger step amplitude should be used to make the noise floor estimate approach the actual noise floor value as quickly as possible.

[0085] Step 4: Reset the threshold to λ = λ ini .

[0086] Step 5: Compare the absolute value of S(k)-N(k-1) with the threshold λ. If the threshold λ is larger, set the correction value e=S(k)-N(k-1); otherwise, set e=sgn(S(k)-N(k-1))λ, where sgn(x) represents the sign of x.

[0087] Step 6: Update the noise floor N(k) to N(k-1)+e, update the power spectrum index value k=k+1, and return to step 2 until the noise floor of the entire power spectrum is estimated, and then denoise.

[0088] Based on the noise floor estimation, the center frequency and bandwidth of multiple signals are estimated using the peak finding method. The specific process is as follows:

[0089] Step 1: Initialize and set the initial value, so that the local minimum and maximum values ​​p min =p max =S(1), power spectrum index value k=1, peak mark f peak , when in the peak range, f peak =1; when in the non-peak range, f peak =0; boundary mark f edge , when at the boundary, f edge =1; when it is not on the boundary, f edge =0.

[0090] Step 2: If S(k)>p max And f peak =1, indicating that the current power spectrum value is still within the signal peak range, then the local maximum value p is updated max =S(k).

[0091] Step 3: If S(k) <p min And f peak =0, indicating that the next peak signal has not yet appeared, and the local minimum value p is updated min =S(k).

[0092] Step 4: If S(k) <p max / 4 and f peak =1, that is, the current power spectrum value is less than the local maximum value by 6dB, which means that the current peak range has been terminated, and the signal termination boundary is to be found. Let f peak =0,f edge =1.

[0093] Here, 1 / 4 of the maximum power is equivalent to a 6dB power drop.

[0094] Step 5: If S(k)>4p min And f peak = 0, that is, the index k is equal to the number of power spectrum values ​​M and f edge =1, it indicates that the previous signal has ended, and the current index value k is recorded as the termination boundary of the current signal.

[0095] Update the signal boundary range, starting from the local maximum, traversing forward and backward until the signal is lower than the noise floor for the first time, and use the corresponding values ​​as the new starting and ending boundaries of the signal.

[0096] The bandwidth of the signal is:

[0097] The power spectrum value corresponding to the end boundary - the power spectrum value corresponding to the start boundary

[0098] The center frequency of the signal is:

[0099] (Power spectrum value corresponding to the end boundary + power spectrum value corresponding to the start boundary) / 2

[0100] Let the peak mark f pedk =1, boundary mark f edge =0, record the current index k as the starting boundary of the next signal.

[0101] Step 6: Update the power spectrum index value to k+1. If k is greater than the number of power spectrum values ​​M, the process ends; otherwise, proceed to step 2.

[0102] Example 2

[0103] A radiation source signal detection method provided by the present invention includes:

[0104] Step S1: Estimating the noise floor of the received radiation source signal and denoising it;

[0105] Step S2: estimating the center frequency and bandwidth of the denoised signal;

[0106] Step S3: extract the center frequency and bandwidth of the signal as the detection signal.

[0107] Specifically, step S1 includes:

[0108] Step S1.1: The collected power spectrum is S, the power spectrum value threshold is λ, and the original threshold is set to λ ini , downward adjustment coefficient a, downward adjustment coefficient b;

[0109] Where k is the index value of the power spectrum, a<1, b>1; the kth noise floor estimation value is N(k), and S(k) is the kth power spectrum value;

[0110] Step S1.2: Determine the sign of (S(k)-N(k-1))×(S(k)-N(k-2)). If the sign is positive, execute step S1.3; if the sign is negative, execute step S1.4;

[0111] Step S1.3: Determine the sign of S(k)-N(k-1). If the sign is positive, update the threshold to λ. New =λ×a; if the sign is negative, update the threshold to λ New =λ×b; after updating the threshold, execute step S1.5;

[0112] Step S1.4: Let the threshold λ New =λ ini , then execute step S1.5;

[0113] Step S1.5: Compare the absolute value of S(k)-N(k-1) with the threshold λ New , if the threshold λ New If the threshold λ is large, let the correction value e = S(k)-N(k-1); New If it is small, let e = sgn(S(k)-N(k-1))λ New , where sgn(x) represents the sign of x;

[0114] Step S1.6: Update the noise floor N(k)=N(k-1)+e, update the index value to k+1, return to step S1.2, and execute step S1.7 after estimating the noise floor of the entire power spectrum;

[0115] Step S1.7: In the collected power spectrum, remove the noise floor to complete denoising.

[0116] Specifically, step S2 includes:

[0117] Step S2.1: Set the definition rule of the mark value, the peak mark is f peak , when in the peak range, f peak =1; when in the non-peak range, f peak =0; the boundary mark is f edge , when at the boundary, f edge =1; when it is not on the boundary, f edge =0;

[0118] Step S2.2: Determine the minimum value p of the local power spectrum value min and the maximum value p max ; Specifically, when S(k)>p max And f peak =1, the local maximum value is updated to p max =S(k), otherwise p is not updated max ; When S(k) <p min And f peak = 0, update the local minimum value to p min =S(k), otherwise p is not updated min ;

[0119] Step S2.3: Estimate the start boundary and the end boundary of the signal according to the maximum and minimum values ​​of the local power spectrum values;

[0120] Step S2.4: Calculate the bandwidth and center frequency of the signal through the power spectrum value corresponding to the boundary, and set the peak mark fpeak =1, boundary mark f edge =0;

[0121] Signal bandwidth = power spectrum value corresponding to the end boundary - power spectrum value corresponding to the start boundary

[0122] The center frequency of the signal = (the power spectrum value corresponding to the end boundary + the power spectrum value corresponding to the start boundary) / 2;

[0123] Step S2.5: Update the power spectrum index value to k+1. When the power index value is greater than the number of power spectrum values ​​M, proceed to step S3; otherwise return to execute steps S2.2-S2.4 until the power index value is greater than the number of power spectrum values ​​M.

[0124] Specifically, step S2.3 includes:

[0125] Step S2.3.1: If S(k) <p max / 4 and f peak =1, let f peak =0,f edge =1, proceed to step S2.3.2; otherwise, proceed directly to step S2.3.3;

[0126] Step S2.3.2: If S(k)>4p min And f peak = 0, or the index value k is equal to the number of spectral values ​​M of the power spectrum and f edge =1, it indicates that the previous signal ends, and the current index value k is recorded as the current signal end boundary End1; otherwise, proceed to step S2.5;

[0127] Step S2.3.3: Update the signal boundary range; starting from the local maximum value, traverse the power spectrum value forward and backward, and when the power spectrum value of the signal is lower than the background noise for the first time, use the index value corresponding to the power spectrum value as the starting boundary Start and ending boundary End2 of the signal;

[0128] Step S2.3.4: Determine the final termination boundary End; compare the power spectrum values ​​corresponding to the termination boundary End1 and the termination boundary End2, take the termination boundary with the larger power spectrum value as the final termination boundary End, and set the termination boundary End as the starting boundary of the next signal.

[0129] Specifically, step S3 includes:

[0130] The central frequency and bandwidth parts of the radiation source signal are extracted, and the other parts of the signal are filtered out to obtain the detection signal.

[0131] For each signal, based on its estimated center frequency and bandwidth results, a digital down-conversion filter with different parameters is selected to select the signal and filter out other signals, and further estimate the time domain parameters of the signal.

[0132] The present invention also provides a radiation source signal detection system, which can be implemented by executing the process steps of the radiation source signal detection method, that is, those skilled in the art can understand the radiation source signal detection method as a preferred implementation of the radiation source signal detection system.

[0133] A radiation source signal detection system provided according to the present invention includes: module M1: estimating the noise floor of the received radiation source signal and denoising it; module M2: estimating the center frequency and bandwidth of the denoised signal; module M3: extracting the center frequency and bandwidth parts of the signal as a detection signal.

[0134] Specifically, module M1 includes: Module M1.1: The collected power spectrum is S, the power spectrum value threshold is λ, and the original threshold is set to λ ini , down-adjustment coefficient a, down-adjustment coefficient b; where k is the index value of the power spectrum, a<1, b>1; the kth noise floor estimation value is N(k), and S(k) is the kth power spectrum value; module M1.2: determine the sign of (S(k)-N(k-1))×(S(k)-N(k-2)), if the sign is positive, execute module M1.3; if the sign is negative, execute module M1.4; module M1.3: determine the sign of S(k)-N(k-1), if the sign is positive, update the threshold to λ New =λ×a; if the sign is negative, update the threshold to λ New =λ×b; After updating the threshold, execute module M1.5; Module M1.4: Set the threshold λ New =λ ini , then execute module M1.5; Module M1.5: Compare the absolute value of S(k)-N(k-1) with the threshold λ New , if the threshold λ New If the threshold λ is large, let the correction value e = S(k)-N(k-1); New If it is small, let e = sgn(S(k)-N(k-1))λ New , where sgn(x) represents the sign of x; Module M1.6: Update the noise floor N(k) = N(k-1) + e, update the index value to k+1, return to execute module M1.2, and execute module M1.7 after estimating the noise floor of the entire power spectrum; Module M1.7: Remove the noise floor from the collected power spectrum to complete denoising.

[0135] Specifically, module M2 includes: Module M2.1: Setting the definition rule of the mark value, the peak mark is f peak , when in the peak range, f peak =1; when in the non-peak range, f peak =0; the boundary mark is f edge , when at the boundary, f edge =1; when it is not on the boundary, f edge =0; Module M2.2: Determine the minimum value p of the local power spectrum min and the maximum value p max ; Specifically, when S(k)>p max And f peak =1, the local maximum value is updated to p max =S(k), otherwise p is not updated max ; When S(k) <p min And f peak = 0, update the local minimum value to p min =S(k), otherwise p is not updated min Module M2.3: Estimate the start and end boundaries of the signal based on the maximum and minimum values ​​of the local power spectrum; Module M2.4: Calculate the bandwidth and center frequency of the signal through the power spectrum values ​​corresponding to the boundaries, and set the peak mark f peak =1, boundary mark f edge =0;

[0136] Signal bandwidth = power spectrum value corresponding to the end boundary - power spectrum value corresponding to the start boundary

[0137] The center frequency of the signal = (the power spectrum value corresponding to the end boundary + the power spectrum value corresponding to the start boundary) / 2;

[0138] Module M2.5: Update the power spectrum index value to k+1. When the power index value is greater than the number of power spectrum values ​​M, proceed to module M3; otherwise return to execute modules M2.2-M2.4 until the power index value is greater than the number of power spectrum values ​​M.

[0139] Specifically, module M2.3 includes: Module M2.3.1: If S(k) <p max / 4 and f peak =1, let f peak =0,f edge =1, proceed to module M2.3.2; otherwise proceed directly to module M2.3.3; Module M2.3.2: If S(k)>4p min And f peak = 0, or the index value k is equal to the number of spectral values ​​M of the power spectrum and f edge=1, it indicates that the previous signal ends, and the current index value k is recorded as the current signal termination boundary End1; otherwise, proceed to module M2.5; module M2.3.3: update the signal boundary range; starting from the local maximum value, traverse the power spectrum value forward and backward, and when the power spectrum value of the signal is lower than the background noise for the first time, use the index value corresponding to the power spectrum value as the starting boundary Start and the ending boundary End2 of the signal; module M2.3.4: determine the final termination boundary End; compare the power spectrum values ​​corresponding to the termination boundary End1 and the termination boundary End2, take the termination boundary with the larger power spectrum value as the final termination boundary End, and set the termination boundary End as the starting boundary of the next signal.

[0140] Specifically, module M3 includes: extracting the central frequency and bandwidth parts in the radiation source signal, and filtering out other parts of the signal to obtain a detection signal.

[0141] Those skilled in the art know that, in addition to realizing the system and its various devices, modules, and units provided by the present invention in a purely computer-readable program code, it is entirely possible to realize the same functions in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered as a hardware component, and the devices, modules, and units included therein for realizing various functions can also be regarded as structures within the hardware component; the devices, modules, and units for realizing various functions can also be regarded as both software modules for realizing the method and structures within the hardware component.

[0142] The above describes the specific embodiments of the present invention. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essence of the present invention. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A radiation source signal detection method, characterized in that: include: Step S1: Estimating the noise floor of the received radiation source signal and denoising it; Step S2: estimating the center frequency and bandwidth of the denoised signal; Step S3: extract the center frequency and bandwidth parts in the signal as the detection signal.

2. The radiation source signal detection method according to claim 1, characterized in that: Step S1 includes: Step S1.1: The collected power spectrum is S, the power spectrum value threshold is λ, and the original threshold is set to λ ini , downward adjustment coefficient a, downward adjustment coefficient b; Where k is the index value of the power spectrum, a<1, b>1; the k-th noise floor estimate is N(k), and S(k) is the k-th power spectrum value; Step S1.2: Determine the sign of (S(k)-N(k-1))×(S(k)-N(k-2)). If the sign is positive, execute step S1.3; if the sign is negative, execute step S1.4; Step S1.3: Determine the sign of S(k)-N(k-1). If the sign is positive, update the threshold to λ. New =λ×a; if the sign is negative, update the threshold to λ New =λ×b; after updating the threshold, execute step S1.5; Step S1.4: Let the threshold λ New =λ ini , then execute step S1.5; Step S1.5: Compare the absolute value of S(k)-N(k-1) with the threshold λ New , if the threshold λ New If the threshold λ is large, let the correction value e = S(k)-N(k-1); New If it is small, let e = sgn(S(k)-N(k-1))λ New , where sgn(x) represents the sign of x; Step S1.6: Update the noise floor N(k)=N(k-1)+e, update the index value to k+1, return to step S1.2, and execute step S1.7 after estimating the noise floor of the entire power spectrum; Step S1.7: In the collected power spectrum, remove the noise floor to complete denoising.

3. The radiation source signal detection method according to claim 1, characterized in that: Step S2 includes: Step S2.1: Set the definition rule of the mark value, the peak mark is f peak , when in the peak range, f peak =1; when in the non-peak range, f peak =0; the boundary mark is f edge , when at the boundary, f edge =1; when it is not on the boundary, f edge =0; Step S2.2: Determine the minimum value p of the local power spectrum value min and the maximum value p max ; Specifically, when S(k)>p max And f peak =1, the local maximum value is updated to p max =S(k), otherwise p is not updated max ; When S(k) <p min And f peak = 0, update the local minimum value to p min =S(k), otherwise p is not updated min ; Step S2.3: Estimate the start boundary and the end boundary of the signal according to the maximum and minimum values ​​of the local power spectrum values; Step S2.4: Calculate the bandwidth and center frequency of the signal through the power spectrum value corresponding to the boundary, and set the peak mark f peak =1, boundary mark f edge =0; Signal bandwidth = power spectrum value corresponding to the end boundary minus power spectrum value corresponding to the start boundary The center frequency of the signal = (the power spectrum value corresponding to the end boundary + the power spectrum value corresponding to the start boundary) / 2; Step S2.5: Update the power spectrum index value to k+1. When the power index value is greater than the number of power spectrum values ​​M, proceed to step S3; otherwise return to execute steps S2.2-S2.4 until the power index value is greater than the number of power spectrum values ​​M.

4. The radiation source signal detection method according to claim 3, characterized in that: Step S2.3 includes: Step S2.3.1: If S(k) <p max / 4 and f peak =1, let f peak =0,f edge =1, proceed to step S2.3.2; otherwise, proceed directly to step S2.3.3; Step S2.3.2: If S(k)>4p min And f peak = 0, or the index value k is equal to the number of spectral values ​​M of the power spectrum and f edge =1, it indicates that the previous signal ends, and the current index value k is recorded as the current signal end boundary End1; otherwise, proceed to step S2.5; Step S2.3.3: Update the signal boundary range; starting from the local maximum value, traverse the power spectrum value forward and backward, and when the power spectrum value of the signal is lower than the background noise for the first time, use the index value corresponding to the power spectrum value as the starting boundary Start and ending boundary End2 of the signal; Step S2.3.4: Determine the final termination boundary End; compare the power spectrum values ​​corresponding to the termination boundary End1 and the termination boundary End2, take the termination boundary with the larger power spectrum value as the final termination boundary End, and set the termination boundary End as the starting boundary of the next signal.

5. The radiation source signal detection method according to claim 1, characterized in that: Step S3 includes: The central frequency and bandwidth parts of the radiation source signal are extracted, and the other parts of the signal are filtered out to obtain the detection signal.

6. A radiation source signal detection system, characterized in that: include: Module M1: estimates the noise floor of the received radiation source signal and performs denoising; Module M2: Estimate the center frequency and bandwidth of the denoised signal; Module M3: extract the center frequency and bandwidth of the signal as the detection signal.

7. The radiation source signal detection system according to claim 6, characterized in that: Module M1 includes: Module M1.1: The collected power spectrum is S, the power spectrum value threshold is λ, and the original threshold is set to λ inf , downward adjustment coefficient a, downward adjustment coefficient b; Where k is the index value of the power spectrum, a<1, b>1; the k-th noise floor estimate is N(k), and S(k) is the k-th power spectrum value; Module M1.2: Determine the sign of (S(k)-N(k-1))×(S(k)-N(k-2)). If the sign is positive, execute module M1.3; if the sign is negative, execute module M1.

4. Module M1.3: Determine the sign of S(k)-N(k-1). If the sign is positive, update the threshold to λ New =λ×a; if the sign is negative, update the threshold to λ New =λ×b; after updating the threshold, execute module M1.5; Module M1.4: Let the threshold λ New =λ ini , then execute module M1.5; Module M1.5: Compare the absolute value of S(k)-N(k-1) with the threshold λ New , if the threshold λ New If the threshold λ is large, let the correction value e = S(k)-N(k-1); New If it is small, let e = sgn(S(k)-N(k-1))λ New , where sgn(x) represents the sign of x; Module M1.6: Update the noise floor N(k)=N(k-1)+e, update the index value to k+1, return to execute module M1.2, and execute module M1.7 after estimating the noise floor of the entire power spectrum; Module M1.7: Remove the noise floor from the acquired power spectrum to complete denoising.

8. The radiation source signal detection system according to claim 6, characterized in that: Module M2 includes: Module M2.1: Setting the definition rules of the marker value, the peak marker is f peak , when in the peak range, f peak =1; when in the non-peak range, f peak =0; the boundary mark is f edge , when at the boundary, f edge =1; when it is not on the boundary, f edge =0; Module M2.2: Determine the minimum value p of the local power spectrum min and the maximum value p max ; Specifically, when S(k)>p max And f peak =1, the local maximum value is updated to p max =S(k), otherwise p is not updated max ; When S(k) <p min And f peak = 0, update the local minimum value to p min =S(k), otherwise p is not updated min ; Module M2.3: Estimate the start and end boundaries of the signal based on the maximum and minimum values ​​of the local power spectrum; Module M2.4: Calculate the bandwidth and center frequency of the signal through the power spectrum value corresponding to the boundary, and set the peak mark f peak =1, boundary mark f edge =0; Signal bandwidth = power spectrum value corresponding to the end boundary minus power spectrum value corresponding to the start boundary The center frequency of the signal = (the power spectrum value corresponding to the end boundary + the power spectrum value corresponding to the start boundary) / 2; Module M2.5: Update the power spectrum index value to k+1. When the power index value is greater than the number of power spectrum values ​​M, proceed to module M3; otherwise return to execute modules M2.2-M2.4 until the power index value is greater than the number of power spectrum values ​​M.

9. The radiation source signal detection system according to claim 8, characterized in that: Module M2.3 includes: Module M2.3.1: If S(k) <p max / 4 and f peak =1, let f peak =0,f edge =1, proceed to module M2.3.2; otherwise proceed directly to module M2.3.3; Module M2.3.2: If S(k)>4p min And f peak = 0, or the index value k is equal to the number of spectral values ​​M of the power spectrum and f edge =1, it indicates that the previous signal ends, and the current index value k is recorded as the current signal termination boundary End1; otherwise, proceed to module M2.5; Module M2.3.3: Update the signal boundary range; starting from the local maximum value, traverse the power spectrum value forward and backward. When the power spectrum value of the signal is lower than the background noise for the first time, the index value corresponding to the power spectrum value is used as the starting boundary Start and ending boundary End2 of the signal; Module M2.3.4: Determine the final termination boundary End; compare the power spectrum values ​​corresponding to the termination boundary End1 and the termination boundary End2, take the termination boundary with the larger power spectrum value as the final termination boundary End, and set the termination boundary End as the starting boundary of the next signal.

10. The radiation source signal detection system according to claim 6, characterized in that: Module M3 includes: The central frequency and bandwidth parts of the radiation source signal are extracted, and the other parts of the signal are filtered out to obtain the detection signal.

Images