A modulation signal detection method for a signal receiver

Abstract

The present invention provides a modulation signal detection method for a signal receiver, comprising the following steps: a signal detection module performs smooth filtering on the first time-frequency spectrum data in sequence to obtain the second time-frequency two-dimensional spectrum data; then the noise floor of the second time-frequency two-dimensional spectrum data is estimated to obtain the mean of the time-frequency two-dimensional spectrum data; according to the mean of the time-frequency two-dimensional spectrum data, a decision threshold is calculated to make a signal decision on the second time-frequency two-dimensional spectrum data to obtain a decision matrix for describing whether a signal exists; the average value of the decision matrix on the time frame is binarized to obtain a decision vector F(k); the existence of a target signal is judged by the element value of the binary element in the decision vector F(k); and the target signal parameters are calculated according to the position of the binary element in the decision vector F(k). The present invention solves the technical problem that the target signal is not easy to detect directly and the estimation accuracy of the center frequency and bandwidth of the target signal is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of communication signal detection, and in particular to a modulation signal detection method of a signal receiver. Background Art

[0002] Modulation signals include analog modulation signals and digital modulation signals, among which digital modulation signals include QAM signals, PSK signals, FSK signals, etc. In the field of signal receiver technology, since the generation and transmission of communication signals will be interfered by other signals and communication channels and other unstable factors, the waveform of the received signal will be unstable. Therefore, the modulation signal must be detected first to determine whether there is a target signal in the received data, and then the modulation signal must be identified to determine the modulation type of the signal, so as to facilitate subsequent demodulation, analysis and other signal processing work.

[0003] At present, in the process of detecting modulated signals, there are the following problems: (1) In a complex electromagnetic environment, since the signals in the real environment are usually submerged by noise and other interference signals, the target signal is not easy to detect directly, which may lead to missed detection or false detection of the signal, affecting the communication quality and system performance of the receiver and the inability of the receiver to effectively analyze and process the signal. Therefore, how to accurately determine whether the target signal exists is a technical problem that needs to be solved; (2) Due to the complex and changeable spectrum environment, the dynamic changes of signal frequency and bandwidth are large, and the estimation accuracy of the center frequency and bandwidth of the target signal is reduced, resulting in unreasonable allocation of spectrum resources, increased spectrum interference, and reduced system efficiency. Therefore, how to improve the estimation accuracy of the center frequency and bandwidth of the target signal is another technical problem that needs to be solved at present; (3) Due to the uncertainty of the environmental noise level, directly setting a fixed detection threshold may lead to false detection or missed detection, and may also lead to a large number of false alarms or missed alarms, affecting the reliability and accuracy of signal detection. Therefore, how to accurately estimate the noise level (noise floor) to set the detection threshold is the third problem that needs to be solved. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a modulation signal detection method for a signal receiver, which can detect modulation signals such as QAM signals, PSK signals, FSK signals, etc., and solves the technical problems that the target signal is not easy to detect directly and the estimation accuracy of the center frequency and bandwidth of the target signal is reduced.

[0005] The technical solution of the present invention is as follows:

[0006] A modulation signal detection method for a signal receiver. The signal receiver comprises a signal conditioning module, an FPGA chip communicatively connected to the signal conditioning module, and a processor communicatively connected to the FPGA chip.

[0007] The signal conditioning module is used to filter, gain adjust, correct and amplify the received shortwave signal, and then output the intermediate frequency analog signal to the analog-to-digital conversion module; the FPGA chip is used to receive the intermediate frequency digital signal and convert it into time-frequency two-dimensional spectrum data and then output it to the processor, which includes an analog-to-digital conversion module for converting the intermediate frequency analog signal into an intermediate frequency digital signal, a DC removal module for eliminating the DC component in the intermediate frequency digital signal, and a time-frequency conversion module for converting the intermediate frequency digital signal into time-frequency two-dimensional spectrum data; the processor is used to perform signal detection on the received time-frequency two-dimensional spectrum data; it includes a data acquisition module; the modulation signal detection method includes the following steps:

[0008] After the data acquisition module receives the intermediate frequency digital signal, the DC removal module eliminates the DC component in the intermediate frequency digital signal, and the time-frequency conversion module converts the intermediate frequency digital signal into the first time-frequency two-dimensional spectrum data and outputs it to the processor;

[0009] Acquire first time-frequency two-dimensional spectrum data, and perform smoothing filtering on the first time-frequency two-dimensional spectrum data in sequence to obtain second time-frequency two-dimensional spectrum data; then estimate the noise floor of the second time-frequency two-dimensional spectrum data to obtain the mean value of the time-frequency two-dimensional spectrum data;

[0010] According to the mean value of the time-frequency two-dimensional spectrum data, a decision threshold value is calculated; according to the decision threshold value, a signal decision is made on the second time-frequency two-dimensional spectrum data to obtain a decision matrix for describing whether a signal exists or not;

[0011] The average value of the decision matrix in the time frame is binarized to obtain a decision vector F(k); the decision vector F(k) is composed of a multi-dimensional column vector, each of which contains a plurality of binary elements for describing whether the target signal exists or not, and the dimension of the column vector is the number of frequency frames of the first time-frequency two-dimensional spectrum data;

[0012] The existence of a target signal is determined by the element value of the binary element in the decision vector F(k). If it is detected that the element values ​​of all the binary elements in the decision vector F(k) are 0, it is determined that the target signal is not detected in the first time-frequency two-dimensional spectrum data; if there are one or more binary elements in the decision vector F(k) and the element value is 1, it is determined that the target signal is detected in the first time-frequency two-dimensional spectrum data, wherein the target signal is one or more of an analog modulation signal, a digital modulation signal or a single carrier signal;

[0013] According to the position of the binary elements in the decision vector F(k), the target signal parameters are calculated, and the target signal parameters include the signal center frequency, the signal bandwidth estimation value and the signal quantity.

[0014] Further, obtaining first time-frequency two-dimensional spectrum data, and performing smoothing filtering on the first time-frequency two-dimensional spectrum data to obtain second time-frequency two-dimensional spectrum data specifically includes:

[0015] Obtain any time-frequency spectrum value and its corresponding time frame and frequency frame in the first time-frequency two-dimensional spectrum data, and use the time-frequency spectrum value as the current time-frequency spectrum value and its corresponding time frame as the current time frame;

[0016] Accumulate multiple time-frequency spectrum values ​​corresponding to multiple time frames before the current time frame and multiple time frames after the current time frame at the current frequency frame and divide the sum by the filter length to obtain a smoothing filter value of the current time-frequency spectrum value;

[0017] Each time-frequency spectrum value in the first time-frequency two-dimensional spectrum data is processed according to the above method to obtain smoothing filter values ​​of all time-frequency spectrum values ​​in the first time-frequency two-dimensional spectrum data, and multiple smoothing filter values ​​constitute the second time-frequency two-dimensional spectrum data.

[0018] Further, the noise floor of the second time-frequency two-dimensional spectrum data is estimated to obtain the mean value of the time-frequency two-dimensional spectrum data, which specifically includes:

[0019] Obtain all time frames in the second time-frequency two-dimensional spectrum data, and sequentially accumulate the time-frequency spectrum values ​​corresponding to each time frame to obtain the first time-frequency spectrum value;

[0020] Acquire all frequency frames in the second time-frequency two-dimensional spectrum data, and sequentially accumulate the time-frequency spectrum value corresponding to each frequency frame to obtain a second time-frequency spectrum value;

[0021] The first time-frequency spectrum value and the second time-frequency spectrum value are added and divided by the total number of frames to obtain the mean of the time-frequency two-dimensional spectrum data. The total number of frames is the product of the total number of time frames and the total number of frequency frames.

[0022] Further, according to the mean of the time-frequency two-dimensional spectrum data, a decision threshold is calculated; according to the decision threshold, a signal decision is made on the second time-frequency two-dimensional spectrum data to obtain a decision matrix for describing whether a signal exists or not, which specifically includes:

[0023] Obtain the mean of the time-frequency two-dimensional spectrum data, multiply the mean of the time-frequency two-dimensional spectrum data by the preset value of the hyperparameter, and obtain the decision threshold value;

[0024] Obtaining time-frequency spectrum values ​​in the second time-frequency two-dimensional spectrum data;

[0025] When the time-frequency spectrum value is less than or equal to the decision threshold value, the time-frequency spectrum value in the second time-frequency two-dimensional spectrum data is judged as 0; when the time-frequency spectrum value is greater than the decision threshold value, the time-frequency spectrum value in the second time-frequency two-dimensional spectrum data is judged as 1;

[0026] According to the above method, all time-frequency spectrum values ​​in the second time-frequency two-dimensional spectrum data are judged to obtain a decision matrix; the decision matrix is ​​a binary matrix composed of two dimensions of time and frequency and the matrix element values ​​are 0 or 1.

[0027] Furthermore, the average value of the decision matrix in the time frame is binarized to obtain a decision vector F(k), which specifically includes:

[0028] Obtain all time frames in the decision matrix and the matrix element value corresponding to each time frame, add up the matrix element value corresponding to each time frame in turn and divide by the total number of time frames to obtain a first matrix element mean;

[0029] When the mean value of the first matrix element is less than or equal to the preset value of the hyperparameter, the matrix element value in the decision matrix is ​​judged to be 0; when the mean value of the first matrix element is greater than the preset value of the hyperparameter, the matrix element value in the decision matrix is ​​judged to be 1;

[0030] According to the above method, all matrix element values ​​in the decision matrix are judged to obtain a decision vector F(k). The decision vector F(k) is an N-dimensional column vector whose element values ​​are 0 or 1.

[0031] Further, if there are one or more binary elements in the decision vector F(k) and the element value is 1, it is determined that the target signal is detected in the first time-frequency two-dimensional spectrum data, including:

[0032] If there is a binary element in the decision vector F(k) and the element value is 1, and the other binary elements are 0, it is determined that 1 target signal is detected;

[0033] If there are multiple binary elements with element values ​​of 1 in the decision vector F(k), and the multiple binary elements form one or more element segments, it is determined that one or more target signals are detected, wherein each element segment includes multiple binary elements arranged continuously, and each element segment contains target signals of the same type.

[0034] Further, according to the position of the binary element in the decision vector F(k), the target signal parameters are calculated, specifically including:

[0035] Obtain the frequency frame number k of the binary element whose element value is 1 in the decision vector F(k);

[0036] The frequency frame number k plus 1 is divided by the total number of frequency frames, and then multiplied by half of the signal sampling rate to obtain an estimated value of the target signal center frequency, where the signal sampling rate is the signal sampling rate when the signal is sampled after channelization processing;

[0037] Divide the above signal sampling rate by twice the total number of frequency frames to obtain an estimate of the target signal bandwidth.

[0038] The modulation signal detection method of the signal receiver described in the present invention realizes accurate detection of the target signal by performing signal judgment on the second time-frequency two-dimensional spectrum data, binarizing the average value of the decision matrix on the time frame, and judging the element value of the binary element in the decision vector. It can not only estimate the number, center frequency and bandwidth of the signal, but also improve the spectrum utilization efficiency, optimize the performance of the communication system, enhance the anti-interference capability, support dynamic spectrum management, and assist in signal analysis and classification, and solve key problems such as signal existence detection, signal frequency and bandwidth estimation, and noise baseline estimation, thereby improving the performance and reliability of the overall communication system.

[0039] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0040] 1. The present invention realizes the conversion of externally received shortwave signals into intermediate frequency analog signals and intermediate frequency digital signals into first time-frequency two-dimensional spectrum data and finally performs signal detection on the first time-frequency two-dimensional spectrum data through a signal conditioning module, an FPGA chip connected to the signal conditioning module in communication, and a processor connected to the FPGA chip in communication.

[0041] 2. The present invention performs signal judgment on the second time-frequency two-dimensional spectrum data to obtain a judgment matrix; then binarizes the average value of the judgment matrix on the time frame to obtain a judgment vector, and judges whether a target signal exists by the element value of the binary element in the judgment vector, and calculates the target signal parameters according to the position of the binary element in the judgment vector, thereby achieving the estimation of parameters such as the signal center frequency, the signal bandwidth estimation value and the number of signals, and effectively improving the accuracy of signal parameter estimation, and solving the technical problems existing in the prior art that the spectrum environment is complex and changeable, the dynamic changes of signal frequency and bandwidth are large, the estimation accuracy of the center frequency and bandwidth of the target signal is reduced, resulting in unreasonable allocation of spectrum resources, increased spectrum interference, and reduced system efficiency, and the signals in the real environment are usually submerged by noise and other interference signals, the target signal is not easy to detect directly, which may cause signal leakage or false detection, affecting communication quality and system performance.

[0042] 3. The present invention obtains the mean of the time-frequency two-dimensional spectrum data by adding the first time-frequency spectrum value and the second time-frequency spectrum value and dividing the result by the total number of frames. Then, the decision threshold value is calculated based on the mean of the time-frequency two-dimensional spectrum data, thereby achieving accurate estimation of the noise level (noise floor) to set the detection threshold. This solves the technical problem in the prior art that directly setting a fixed detection threshold may lead to false detection or missed detection due to the uncertainty of the environmental noise level, and may also lead to a large number of false alarms or missed alarms, affecting the reliability and accuracy of signal detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0044] Figure 1 is a schematic diagram of a signal receiver according to an embodiment of the present invention;

[0045] Figure 2 It is a flow chart of a signal detection method of a signal receiver according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The embodiments of the present invention are described in detail below with reference to the accompanying drawings, but the present invention can be implemented in many different ways as defined and covered by the claims.

[0047] As attached Figure 1 As shown, a modulation signal detection method of a signal receiver is mainly used for detecting target signals such as QAM signals, PSK signals, FSK signals, etc.; wherein the modulation signal includes an analog modulation signal, a digital modulation signal or a single carrier signal, and the signal receiver includes a signal conditioning module, an FPGA chip communicatively connected to the signal conditioning module, and a processor communicatively connected to the FPGA chip.

[0048] The signal conditioning module is used to filter, adjust the gain, correct and amplify the received shortwave signal, and then output the intermediate frequency analog signal to the analog-to-digital conversion module; the FPGA chip is used to receive the intermediate frequency digital signal and convert it into time-frequency two-dimensional spectrum data and then output it to the processor, which includes an analog-to-digital conversion module for converting the intermediate frequency analog signal into an intermediate frequency digital signal, a DC removal module for eliminating the DC component in the intermediate frequency digital signal, and a time-frequency conversion module for converting the intermediate frequency digital signal into time-frequency two-dimensional spectrum data; the processor is used to perform signal detection on the received time-frequency two-dimensional spectrum data.

[0049] In this embodiment, when performing signal detection, the target signals that need to be detected are analog modulated signals, digital modulated signals and single-carrier signals; among them, analog modulated signals transmit information by modulating the amplitude, frequency or phase of the carrier, and when performing signal detection, the characteristics of the modulated signal near the carrier frequency appear on the spectrum; digital modulated signals transmit information by modulating discrete changes in the carrier, and common ones include QAM signals, PSK signals, FSK signals, etc. When performing signal detection, they appear as discrete frequency components on the spectrum, and there may be significant signals at multiple frequency points; single-carrier signals, single-carrier signals are concentrated at a specific frequency, and when performing signal detection, they appear as one or more concentrated frequency components on the spectrum.

[0050] As attached Figure 2As shown, in this embodiment, the signal detection method includes the following steps:

[0051] S01. After the data acquisition module receives the intermediate frequency digital signal, the DC removal module eliminates the DC component in the intermediate frequency digital signal, and the time-frequency conversion module converts the intermediate frequency digital signal into the first time-frequency two-dimensional spectrum data and outputs it to the processor.

[0052] S02. Obtain first time-frequency two-dimensional spectrum data, and perform smoothing filtering on the first time-frequency two-dimensional spectrum data in sequence to obtain second time-frequency two-dimensional spectrum data; then estimate the noise floor of the second time-frequency two-dimensional spectrum data to obtain the mean value of the time-frequency two-dimensional spectrum data.

[0053] In this step, the first time-frequency two-dimensional spectrum data and the second time-frequency two-dimensional spectrum data are both sets of multiple time-frequency spectrum values, each of which corresponds to a time frame and a frequency frame. The first time-frequency two-dimensional spectrum data is a function of time and frequency, and can reflect the characteristics of the signal changing with time and frequency. Its expression is S x (i, k), i is the time frame number, i=1,2,3...Q, k is the frequency frame number, k=1,2,3...N-1. The first time-frequency two-dimensional spectrum data is generally obtained by obtaining a frame of first time-frequency two-dimensional spectrum data.

[0054] Preferably, in step S02, first time-frequency two-dimensional spectrum data is obtained, and second time-frequency two-dimensional spectrum data is obtained after smoothing filtering of the first time-frequency two-dimensional spectrum data, which specifically includes:

[0055] S201. Obtain any time-frequency spectrum value and its corresponding time frame and frequency frame in the first time-frequency two-dimensional spectrum data, and use the any time-frequency spectrum value as the current time-frequency spectrum value and its corresponding time frame as the current time frame.

[0056] S202. Accumulate the multiple time-frequency spectrum values ​​corresponding to the multiple time frames before the current time frame and the multiple time frames after the current time frame at the current frequency frame and divide the sum by the filter length to obtain the smoothing filter value of the current time-frequency spectrum value; wherein the multiple time frames are (L-1) / 2 time frames, and L is the length of the smoothing filter, that is, the number of frequency frames considered when performing smoothing processing in the frequency dimension.

[0057] S203. Process each time-frequency spectrum value in the first time-frequency two-dimensional spectrum data according to the above method to obtain smoothing filter values ​​of all time-frequency spectrum values ​​in the first time-frequency two-dimensional spectrum data, and multiple smoothing filter values ​​constitute the second time-frequency two-dimensional spectrum data.

[0058] The above steps S201 to S203 are steps for smoothing and filtering the first time-frequency two-dimensional spectrum data in this embodiment, the purpose of which is to reduce the influence of noise and make the characteristics of the signal more obvious; specifically, by averaging the first time-frequency two-dimensional spectrum data in the frequency dimension, the influence of high-frequency noise is reduced, thereby highlighting the main components of the signal. In this embodiment, the smoothing filter is performed in the frequency dimension, and the expression of the second time-frequency two-dimensional spectrum data after smoothing filtering is:

[0059]

[0060] In formula (1), S x (j, k) is the first time-frequency two-dimensional spectrum data, representing the time-frequency spectrum value at time frame j and frequency frame k; S y (i, k) is the second time-frequency two-dimensional spectrum data, L is the length of the filter, j is the time frame number range used for smoothing filtering, and its minimum value is The maximum value is

[0061] Preferably, in step S02, estimating the noise floor of the second time-frequency two-dimensional spectrum data to obtain the mean of the time-frequency two-dimensional spectrum data specifically includes:

[0062] S204. Obtain all time frames in the second time-frequency two-dimensional spectrum data, and sequentially accumulate the time-frequency spectrum values ​​corresponding to each time frame to obtain the first time-frequency spectrum value.

[0063] S205. Obtain all frequency frames in the second time-frequency two-dimensional spectrum data, and sequentially accumulate the time-frequency spectrum value corresponding to each frequency frame to obtain a second time-frequency spectrum value.

[0064] S206. Add the first time-frequency spectrum value and the second time-frequency spectrum value and divide the sum by the total number of frames to obtain a mean value of the time-frequency two-dimensional spectrum data, where the total number of frames is the product of the total number of time frames and the total number of frequency frames.

[0065] The purpose of the above steps S204 to S206 is to estimate the noise floor for improving signal detection, recognition and enhancement. The noise floor (referred to as "noise floor") refers to the average level of background noise in the absence of a significant signal. It is a benchmark for determining whether a signal exists and whether its strength is large enough to exceed the background noise. In an embodiment of the present invention, the noise floor σ can be estimated as follows:

[0066]

[0067] In formula (2), S y (j, k) is the second time-frequency two-dimensional spectrum data, N is the total number of frequency frames, Q is the total number of time frames, i is the time frame number, and k is the frequency frame number.

[0068] S03. Calculate the decision threshold value based on the mean value of the time-frequency two-dimensional spectrum data; perform signal decision on the second time-frequency two-dimensional spectrum data based on the decision threshold value to obtain a decision matrix for describing whether the signal exists or not.

[0069] Preferably, in step S03, a decision threshold is calculated based on the mean of the time-frequency two-dimensional spectrum data; and a signal decision is performed on the second time-frequency two-dimensional spectrum data based on the decision threshold to obtain a decision matrix for describing whether a signal exists or not, which specifically includes:

[0070] S301. Obtain the mean of the time-frequency two-dimensional spectrum data, multiply the mean of the time-frequency two-dimensional spectrum data by the preset value of the hyperparameter, and obtain the decision threshold value.

[0071] S302. Obtain time-frequency spectrum values ​​in the second time-frequency two-dimensional spectrum data.

[0072] S303. When the time-frequency spectrum value is less than or equal to the decision threshold value, the time-frequency spectrum value in the second time-frequency two-dimensional spectrum data is judged to be 0; when the time-frequency spectrum value is greater than the decision threshold value, the time-frequency spectrum value in the second time-frequency two-dimensional spectrum data is judged to be 1.

[0073] S304. Judge all the time-frequency spectrum values ​​in the second time-frequency two-dimensional spectrum data according to the above method to obtain a decision matrix; the decision matrix is ​​a binary matrix composed of two dimensions of time and frequency and the matrix element values ​​are 0 or 1.

[0074] In the above step S301, the decision threshold th is calculated according to the following formula:

[0075] th=λσ1 (3)

[0076] In formula (3), σ1 is the mean of the time-frequency two-dimensional spectrum data, and λ is a hyperparameter. The default value is generally 3, which will be adjusted and optimized according to the actual data later.

[0077] The above step S303 is to construct a decision matrix S by making a 0 / 1 decision on the second time-frequency two-dimensional spectrum data after smoothing filtering according to the decision threshold. z (i, k), thereby achieving the purpose of converting the smoothed second time-frequency two-dimensional spectrum data into a binary matrix. z (i,k) is a binary matrix with the same dimensions as Sy(i,k), where 1 indicates the presence of a signal and 0 indicates the absence of a signal. The calculation process can be described as:

[0078]

[0079] In formula (4), S y (i, k) is the second time-frequency two-dimensional spectrum data, th is the decision threshold, i is the time frame number, and k is the frequency frame number.

[0080] S04. Binarize the average value of the decision matrix in the time frame to obtain a decision vector F(k); the decision vector F(k) is composed of a multi-dimensional column vector, each dimension of the column vector contains multiple binary elements for describing whether the target signal exists or not, and the dimension of the column vector is the number of frequency frames of the first time-frequency two-dimensional spectrum data.

[0081] Preferably, in step S04, binarization is performed on the average value of the decision matrix in the time frame to obtain a decision vector F(k), which specifically includes:

[0082] S401. Obtain all time frames in the decision matrix and the matrix element value corresponding to each time frame, sequentially add up the matrix element value corresponding to each time frame in the same frequency frame and divide by the total number of time frames to obtain a first matrix element mean.

[0083] S402. When the mean value of the first matrix elements is less than or equal to the preset threshold, the matrix element value in the decision matrix is ​​judged to be 0; when the mean value of the first matrix elements is greater than the preset value of the hyperparameter, the matrix element value in the decision matrix is ​​judged to be 1.

[0084] S403. Determine all matrix element values ​​in the decision matrix according to the above method to obtain a decision vector F(k), where the decision vector F(k) is an N-dimensional column vector whose element values ​​are 0 or 1.

[0085] In the above steps S401 to S403, the decision vector F(k) is an F-dimensional column vector, where F is the frequency frame number, which is obtained by transforming the decision matrix S z (i, k) is obtained by binarizing the average value of the time frame, each element F(k) is 0 or 1, indicating whether there is a signal in the k-th frequency frame; if the mean value of the first matrix element of the k-th frequency frame in all time frames exceeds or is equal to the preset threshold r, then F(k) = 1, indicating that there is a signal, otherwise F(k) = 0, indicating that there is no signal. Specifically, the k-th element F(k) of the decision vector is a binary value (0 or 1), according to S z The column average of (i, k) is compared with the preset threshold r and can be described as:

[0086]

[0087] In formula (5), S z (i, k) is a decision matrix, which can be described as the matrix element value when the time frame is i and the frequency frame is k; Q is the total number of time frames, and r is a preset threshold.

[0088] S05. Determine whether there is a target signal by the element value of the binary element in the decision vector F(k). If it is detected that the element values ​​of all binary elements in the decision vector F(k) are 0, it is determined that no target signal is detected in the first time-frequency two-dimensional spectrum data; if there are one or more binary elements in the decision vector F(k) and the element value is 1, it is determined that the target signal is detected in the first time-frequency two-dimensional spectrum data.

[0089] Preferably, in step S05, if there are one or more binary elements in the decision vector F(k) and the element value is 1, it is determined that the target signal is detected in the first time-frequency two-dimensional spectrum data, including:

[0090] S501. If there is a binary element in the decision vector F(k) and the element value is 1, and the other binary elements are 0, it is determined that 1 target signal is detected.

[0091] S502. If there are multiple binary elements with element value 1 in the decision vector F(k), and the multiple binary elements form one or more element segments, it is determined that one or more target signals are detected, wherein each element segment includes multiple binary elements arranged continuously, and each element segment contains target signals of the same type.

[0092] S06. According to the position of the binary element in the decision vector F(k), the target signal parameters are calculated, and the target signal parameters include the signal center frequency, the signal bandwidth estimation value and the signal quantity.

[0093] Preferably, in the above step S06, the target signal parameter is calculated according to the position of the binary element in the decision vector F(k), specifically including:

[0094] S601. Obtain the frequency frame number k of the binary element whose element value is 1 in the decision vector F(k);

[0095] S602. Add 1 to the frequency frame number k and divide it by the total number of frequency frames, and then multiply it by half of the signal sampling rate to obtain an estimated value of the target signal center frequency, where the signal sampling rate is the signal sampling rate when the signal is sampled after channelization processing.

[0096] S603. Divide the above signal sampling rate by twice the total number of frequency frames to obtain an estimated value of the target signal bandwidth.

[0097] In the above steps S601 to S603, the frequency frame number of the binary element with an element value of 1 in the decision vector F(k) is set to the jth element, that is, the subscript value of the binary element with an element value of 1 in the decision vector F(k) is set to j, and the estimated value of the center frequency of the target signal can be calculated. and an estimate of the bandwidth of the target signal Among them, the estimated value of the center frequency of the target signal is for:

[0098]

[0099] The estimated bandwidth of the target signal is for:

[0100]

[0101] In formula (6) and formula (7), f s is the sampling rate after channelization processing, and N is the total number of frequency frames.

[0102] The embodiment of the present invention first converts the received intermediate frequency digital signal into first time-frequency two-dimensional spectrum data through a data acquisition module and outputs it to a signal detection module in a processor, then performs smoothing filtering and noise floor estimation on the first time-frequency two-dimensional spectrum data through the signal detection module to obtain the mean value and decision threshold value of the time-frequency two-dimensional spectrum data; then, according to the decision threshold value, a signal decision is made on the second time-frequency two-dimensional spectrum data to obtain a decision matrix for describing whether a signal exists or not; finally, a decision vector is obtained by binarizing the average value of the decision matrix on a time frame, and whether a target signal exists is determined by the element value of the binary element in the decision vector, and the target signal parameters are calculated according to the position of the binary element in the decision vector, thereby achieving the estimation of parameters such as the signal center frequency, the signal bandwidth estimation value and the number of signals, and effectively improving the accuracy of signal parameter estimation.

[0103] At the same time, the present invention also connects the signal conditioning module with the analog-to-digital conversion module in the FPGA chip, and the analog-to-digital conversion module is communicated with the time-frequency conversion module and the signal detection module in turn, so that the signal conditioning module filters, adjusts the gain, corrects and amplifies the received shortwave signal and outputs an intermediate frequency analog signal, and the FPGA chip converts the intermediate frequency analog signal into a first time-frequency two-dimensional spectrum data, and the signal detection module performs signal detection on the first time-frequency two-dimensional spectrum data, and achieves the effect of effectively detecting the target signal and accurately estimating the center frequency and bandwidth of the target signal, solving the technical problems existing in the prior art that the target signal is not easy to be directly detected when the signal is usually submerged by noise and other interference signals, resulting in signal missed detection or false detection, and the technical problem that the estimation accuracy of the center frequency and bandwidth of the target signal is reduced due to the complex and changeable spectrum environment and the large dynamic changes of the signal frequency and bandwidth.

Claims

1. A modulation signal detection method for a signal receiver, the signal receiver comprising a signal conditioning module, an FPGA chip communicatively connected to the signal conditioning module, and a processor communicatively connected to the FPGA chip, characterized in that: The signal conditioning module is used to filter, gain adjust, correct and amplify the received shortwave signal, and then output the intermediate frequency analog signal to the analog-to-digital conversion module; the FPGA chip is used to receive the intermediate frequency digital signal and convert it into time-frequency two-dimensional spectrum data and then output it to the processor, which includes an analog-to-digital conversion module for converting the intermediate frequency analog signal into an intermediate frequency digital signal, a DC removal module for eliminating the DC component in the intermediate frequency digital signal, and a time-frequency conversion module for converting the intermediate frequency digital signal into the first time-frequency two-dimensional spectrum data; The processor is used to perform signal detection on the received first time-frequency two-dimensional spectrum data, and includes a signal detection module; the modulation signal detection method includes the following steps: After the data acquisition module receives the intermediate frequency digital signal, the DC removal module eliminates the DC component in the intermediate frequency digital signal, and the time-frequency conversion module converts the intermediate frequency digital signal into the first time-frequency two-dimensional spectrum data and outputs it to the signal detection module in the processor; The signal detection module obtains the first time-frequency two-dimensional spectrum data, and performs smoothing filtering on the first time-frequency two-dimensional spectrum data in turn to obtain the second time-frequency two-dimensional spectrum data; then the noise floor of the second time-frequency two-dimensional spectrum data is estimated to obtain the mean of the time-frequency two-dimensional spectrum data; the estimating the noise floor of the second time-frequency two-dimensional spectrum data to obtain the mean of the time-frequency two-dimensional spectrum data specifically includes: obtaining all time frames in the second time-frequency two-dimensional spectrum data, and accumulating the time-frequency spectrum values ​​corresponding to each time frame in turn to obtain the first time-frequency spectrum value; obtaining all frequency frames in the second time-frequency two-dimensional spectrum data, and accumulating the time-frequency spectrum values ​​corresponding to each frequency frame in turn to obtain the second time-frequency spectrum value; adding the first time-frequency spectrum value and the second time-frequency spectrum value and dividing the sum by the total number of frames to obtain the mean of the time-frequency two-dimensional spectrum data, wherein the total number of frames is the product of the total number of time frames and the total number of frequency frames; According to the mean value of the time-frequency two-dimensional spectrum data, a decision threshold value is calculated; according to the decision threshold value, a signal decision is made on the second time-frequency two-dimensional spectrum data to obtain a decision matrix for describing whether a signal exists or not; Binarize the average value of the decision matrix on the time frame to obtain a decision vector F(k); the decision vector F(k) is composed of a multi-dimensional column vector, each dimension of the column vector contains a plurality of binary elements for describing whether the target signal exists or not, the dimension of the column vector is the number of frequency frames of the first time-frequency two-dimensional spectrum data, and k is the frequency frame number; Determine whether a target signal exists by the element value of the binary element in the decision vector F(k); if it is detected that the element values ​​of all binary elements in the decision vector F(k) are 0, it is determined that no target signal is detected in the first time-frequency two-dimensional spectrum data; if there are one or more binary elements in the decision vector F(k) and the element value is 1, it is determined that a target signal is detected in the first time-frequency two-dimensional spectrum data, wherein the target signal is one or more of an analog modulation signal, a digital modulation signal or a single carrier signal; According to the position of the binary element in the decision vector F(k), the target signal parameters are calculated, and the target signal parameters include the signal center frequency, the signal bandwidth estimation value and the signal quantity.

2. A modulation signal detection method for a signal receiver as claimed in claim 1, characterized in that: The obtaining of the first time-frequency two-dimensional spectrum data and obtaining the second time-frequency two-dimensional spectrum data after smoothing filtering the first time-frequency two-dimensional spectrum data specifically includes: Step 201. Obtain any time-frequency spectrum value and its corresponding time frame and frequency frame in the first time-frequency two-dimensional spectrum data, and use the time-frequency spectrum value as the current time-frequency spectrum value and its corresponding time frame as the current time frame; Step 202: Accumulate multiple time-frequency spectrum values ​​corresponding to multiple time frames before the current time frame and multiple time frames after the current time frame at the current frequency frame and divide the sum by the filter length to obtain a smoothing filter value of the current time-frequency spectrum value; According to step 201 to step 202, each time-frequency spectrum value in the first time-frequency two-dimensional spectrum data is processed to obtain smoothing filter values ​​of all time-frequency spectrum values ​​in the first time-frequency two-dimensional spectrum data, and multiple smoothing filter values ​​constitute the second time-frequency two-dimensional spectrum data.

3. A modulation signal detection method for a signal receiver as claimed in claim 2, characterized in that: The decision threshold value is calculated based on the mean of the two-dimensional time-frequency spectrum data; According to the decision threshold value, a signal decision is made on the second time-frequency two-dimensional spectrum data to obtain a decision matrix for describing whether a signal exists or not, which specifically includes: Step 301. Obtain the mean of the time-frequency two-dimensional spectrum data, and multiply the mean of the time-frequency two-dimensional spectrum data by the preset value of the hyperparameter to obtain a decision threshold value; Step 302: Obtain time-frequency spectrum values ​​in the second time-frequency two-dimensional spectrum data; Step 303: When the time-frequency spectrum value is less than or equal to the decision threshold value, the time-frequency spectrum value in the second time-frequency two-dimensional spectrum data is judged as 0; when the time-frequency spectrum value is greater than the decision threshold value, the time-frequency spectrum value in the second time-frequency two-dimensional spectrum data is judged as 1; According to step 301 to step 303, all time-frequency spectrum values ​​in the second time-frequency two-dimensional spectrum data are judged to obtain a decision matrix; the decision matrix is ​​a binary matrix composed of two dimensions of time and frequency and the matrix element values ​​are 0 or 1.

4. A modulation signal detection method for a signal receiver as claimed in claim 3, characterized in that: The binarization process of the average value of the decision matrix in the time frame to obtain the decision vector F(k) specifically includes: Step 401. Obtain all time frames in the decision matrix and the matrix element value corresponding to each time frame, add up the matrix element value corresponding to each time frame in sequence and divide by the total number of time frames to obtain the first matrix element mean; Step 402: When the mean value of the first matrix element is less than or equal to the preset value of the hyperparameter, the matrix element value in the decision matrix is ​​judged to be 0; when the mean value of the first matrix element is greater than the preset value of the hyperparameter, the matrix element value in the decision matrix is ​​judged to be 1; According to step 401 to step 402, all matrix element values ​​in the decision matrix are judged to obtain a decision vector F(k), where the decision vector F(k) is an N-dimensional column vector whose element values ​​are 0 or 1.

5. A modulation signal detection method for a signal receiver as claimed in claim 4, characterized in that: If there are one or more binary elements in the decision vector F(k) and the element value is 1, it is determined that the target signal is detected in the first time-frequency two-dimensional spectrum data, including: If there is a binary element in the decision vector F(k) and the element value is 1, and the other binary elements are 0, it is determined that 1 target signal is detected; If there are multiple binary elements with element values ​​of 1 in the decision vector F(k), and the multiple binary elements form one or more element segments, it is determined that one or more target signals are detected, wherein each element segment includes multiple binary elements arranged continuously, and each element segment contains target signals of the same type.

6. A modulation signal detection method for a signal receiver as claimed in claim 5, characterized in that: The target signal parameters are calculated according to the positions of the binary elements in the decision vector F(k), specifically including: Obtain the frequency frame number k of the binary element whose element value is 1 in the decision vector F(k); The frequency frame number k plus 1 is divided by the total number of frequency frames, and then multiplied by half of the signal sampling rate to obtain an estimated value of the target signal center frequency, where the signal sampling rate is the signal sampling rate when the signal is sampled after channelization processing; The signal sampling rate is divided by twice the total number of frequency frames to obtain an estimate of the target signal bandwidth.

Images