A fast signal blind detection synchronization method under low signal-to-noise ratio

Abstract

This invention proposes a fast blind signal detection synchronization method under low signal-to-noise ratio (SNR) conditions. It achieves simultaneous multi-symbol detection and signal demodulation through time-domain processing, resulting in higher real-time performance. By setting different sampling positions within a symbol period to determine the optimal sampling decision time, strict carrier synchronization is not required, reducing timing errors and improving the accuracy of symbol decision. Good demodulation performance is also achieved under incoherent demodulation conditions. The combined use of the decision unit and limiter effectively suppresses the impact of impulse noise on signal detection. This invention improves detection performance by increasing the values ​​of statistical features, allowing for a more intuitive distinction between signal and noise, and is more conducive to signal detection under low SNR conditions. This invention overcomes the shortcomings of existing methods that cannot adequately balance real-time performance and reliability, improving signal detection efficiency and performance, and providing a viable detection scheme for practical applications.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication and signal processing technology, and specifically relates to a fast blind signal detection and synchronization method under low signal-to-noise ratio. Background Technology

[0002] In digital communication systems, because the clocks at the transmitting and receiving ends originate from different local oscillators and are not perfectly synchronized, and because signal delays occur during transmission, the receiver cannot sample at the optimal decision point for each symbol. The presence of filters introduces some signal distortion, which reduces the signal-to-noise ratio of the sampled data and introduces inter-symbol interference (ISI), increasing the bit error rate (BER) during demodulation and degrading the performance of the communication system. Timing recovery is a crucial module that determines the overall operation and performance of the receiver. Its function is to sample at the symbol rate at time t = mT + τ to obtain the transmitted symbols, where T is the symbol interval and τ is the timing error; a smaller timing error results in a lower BER.

[0003] In non-cooperative communication, the receiver, as an unauthorized access user, cannot obtain the modulation parameters of the transmitter. To obtain the information content transmitted by the received signal, it is generally necessary to successfully intercept the signal and obtain its relevant modulation parameters. The prerequisite for intercepting the target signal is to determine whether a communication signal exists in the received data and to determine the start and end times of the target signal through blind detection.

[0004] The advantages of low-frequency signals include strong penetration, slow attenuation, and long propagation distance. However, in some long-distance communication scenarios, which are non-cooperative, the received signal strength is weak due to noise, interference, and channel fading during propagation, increasing the difficulty of detecting low-frequency signals. This highlights a crucial issue: in non-cooperative communication, low-frequency signals cannot be effectively and quickly detected blindly, requiring consideration of both practicality and interference resistance.

[0005] Common signal detection methods include energy detection, cyclostationary detection, eigenvalue detection, and matched filtering detection. Energy detection is the most classic and widely used traditional detection algorithm, offering advantages such as simple implementation, strong real-time performance, and low prior information requirements. However, its performance is poor at low signal-to-noise ratios (SNRs), making it prone to false alarms and missed alarms. Cyclostationary detection utilizes the difference in cyclostationary characteristics between signals and noise, combined with a spectral correlation function, to distinguish between them. This algorithm has stronger noise immunity, but its computational complexity is higher, and its longer detection time cannot meet the requirements of real-time detection. Eigenvalue detection, based on random matrix theory, first calculates the eigenvalues ​​of the covariance matrix and compares the ratio of the maximum to the minimum eigenvalue with a threshold value to determine the presence or absence of a signal. This algorithm minimizes the impact of noise uncertainty on detection performance; however, the detection threshold lacks a specific analytical expression and can only be estimated through extensive experimentation to obtain a reliable threshold value. A matched filter is defined as a filter that maximizes the output SNR when the received SNR is constant; detection methods based on matched filters are called matched filtering detection. This is equivalent to a signal demodulator; when the filter output achieves the maximum signal-to-noise ratio and exceeds a threshold, the presence of a signal can be determined. This algorithm requires short detection time, but it places high demands on prior information about the signal and on phase synchronization.

[0006] As can be seen from the above, among feasible signal detection schemes, those with superior detection performance and low computational complexity that meet real-time detection requirements are particularly important. Of the schemes mentioned above, constrained by real-time performance and computational complexity requirements, and affected by the received signal-to-noise ratio (SNR), the detection effect for low SNR signals is poor, and they cannot adequately meet practical application requirements in terms of system independence, feasibility, and reliability. Summary of the Invention

[0007] To address the aforementioned problems in the existing technology, this invention provides a fast blind signal detection and synchronization method under low signal-to-noise ratio conditions. The technical problem to be solved by this invention is achieved through the following technical solution:

[0008] The present invention provides a fast signal blind detection and synchronization method under low signal-to-noise ratio conditions, comprising:

[0009] Step 1: Receive the wireless analog signal returned from the far field, and perform analog-to-digital conversion, down-conversion, and downsampling on the wireless analog signal to obtain a downsampled signal;

[0010] Step 2: Divide the downsampled signal continuously according to a fixed window length, and make a decision on the sub-signals within the window to restore the original code.

[0011] Each window contains multiple symbols, each original symbol exists in a symbol period, and each symbol period contains multiple sampling positions;

[0012] Step 3: For each original symbol within a window, take each sampling position within each symbol period as the starting sampling position, and take the window length backward to obtain multiple sampling windows corresponding to each window;

[0013] Step 4: Calculate the demodulation signal-to-noise ratio of each sampling window corresponding to each window, select the starting sampling position of the sampling window with the largest demodulation signal-to-noise ratio as the sampling starting point, and obtain the sampling signal in each window by sampling the signal from the sampling starting point;

[0014] Step 5: Compare the demodulation signal-to-noise ratio of the sampling window corresponding to the sampled signal with the detection threshold to determine whether the sampled signal is the target signal;

[0015] Step 6: For the original code corresponding to the target signal in each window and the next window, compare the second half of the original code in the current window with the first half of the original code in the next window to determine whether there is a misalignment in the original code in the next window. If so, correct the original code in the next window to obtain the corrected code of the target signal in each window.

[0016] Step 7: Differential decoding is performed on the corrected symbols of the target signal within each window to obtain the decoding result.

[0017] The beneficial effects of this invention are:

[0018] This invention proposes a fast blind signal detection synchronization method under low signal-to-noise ratio (SNR) conditions. It achieves simultaneous multi-symbol detection and signal demodulation through time-domain processing, resulting in higher real-time performance. By setting different sampling positions within a symbol period to determine the optimal sampling decision time, strict carrier synchronization is not required, reducing timing errors and improving the accuracy of symbol decision. Good demodulation performance is also achieved under incoherent demodulation conditions. The combined use of the decision unit and limiter effectively suppresses the impact of impulse noise on signal detection. This invention improves detection performance by increasing the values ​​of statistical features, allowing for a more intuitive distinction between signal and noise, and is more conducive to signal detection under low SNR conditions. This invention overcomes the shortcomings of existing methods that cannot adequately balance real-time performance and reliability, improving signal detection efficiency and performance, and providing a viable detection scheme for practical applications.

[0019] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0020] Figure 1This is a schematic diagram of a fast signal blind detection and synchronization method under low signal-to-noise ratio provided by the present invention;

[0021] Figure 2 This is a schematic diagram of the decision-making device of the present invention;

[0022] Figure 3 This is a schematic diagram of the limiter of the present invention;

[0023] Figure 4 This is a schematic diagram of setting sampling points within each window of the present invention;

[0024] Figure 5 This is a schematic diagram of the sampling window of the present invention;

[0025] Figure 6 This is a window display diagram of the code element correction within the window of the present invention;

[0026] Figure 7 This is a schematic diagram comparing two sets of data during the symbol correction process within the window of this invention;

[0027] Figure 8 This is a full demodulation signal-to-noise ratio diagram of the test data of this invention;

[0028] Figure 9 This is a schematic diagram of the detection output results of the present invention;

[0029] Figure 10 This is a comparison diagram of the signal detection probability between the method of this invention and other methods;

[0030] Figure 11 It is a bit error rate curve. Detailed Implementation

[0031] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0032] In a fully digital receiver, the sampling clock and receiver reference carrier are independent of the transmitter. The resulting frequency offset, phase offset, and timing errors are compensated for by algorithms. This compensation method can be implemented in software, making it more versatile and easier to integrate. Traditional correlation detection is optimal under Gaussian noise, but in low-frequency communication, signals transmitted over long distances are affected not only by Gaussian noise such as atmospheric noise but also by sudden strong impulse noise. If conventional methods are continued for synchronization or reception, the performance of the communication system will deteriorate.

[0033] In practical applications, the implementation of a complete function is usually broken down into several parts for research. The detection synchronization algorithm considered in this invention is based on the premise that some prior information is known, the timing error is uncertain, and the received signal-to-noise ratio is low. It performs fast and reliable detection synchronization on the acquired signal, completes demodulation and decoding operations, and recovers the data from the transmitting end. The transmission pattern of low-frequency signals is relatively fixed. With the help of some analysis tools, some prior information, such as the signal modulation method, symbol rate, and center frequency, can be extracted first.

[0034] This invention is used in long-distance broadband receivers to record signals across the entire low-frequency band. The received signals mainly include transmitted signals from high-power stations and electromagnetic signals generated by lightning during propagation. Atmospheric noise is the main interference noise in low-frequency communication. The received signal at a certain frequency point in the signal received by the broadband receiver is the sum of the transmitted signal at that frequency and the noise, as shown in the following formula:

[0035] r t =s t +n t

[0036] n t The mean of the expression is 0, and the variance is σ. 2 Gaussian noise represents the sum of additive noise present in the channel, including thermal noise, atmospheric interference, and industrial interference.

[0037] like Figure 1 As shown, the present invention provides a fast signal blind detection and synchronization method under low signal-to-noise ratio conditions, comprising:

[0038] Step 1: Receive the wireless analog signal returned from the far field, and perform analog-to-digital conversion, down-conversion, and downsampling on the wireless analog signal to obtain a downsampled signal;

[0039] In one specific embodiment, step 1 includes:

[0040] Step 11: Receive the wireless analog signal returned from the far field and convert the wireless analog signal into a digital signal;

[0041] It is worth noting that the software-defined radio (SDR) of the receiver requires processing wireless signals in digital form as much as possible. Therefore, the received analog signal needs to be converted into a digital signal, and the sampling method during the conversion should satisfy the sampling theorem. The application scenario of this invention is to first use a receiver to acquire a high-sampling-rate digital signal, and then use a software-defined radio method to complete the signal detection.

[0042] Step 12: Down-convert the digital signal to convert it from a digital signal to a digital baseband signal;

[0043] To demodulate and restore the original digital signal, the received intermediate frequency signal r needs to be... n,fcConverted into digital baseband signal r n ' ,f0 .

[0044] Step 13: Based on the sampling rate of the digital baseband signal and the required sampling rate, downsample the digital baseband signal to obtain a downsampled signal.

[0045] Specifically, in one optional embodiment, step 13 includes:

[0046] Step 131: Calculate the ratio of the sampling rate of the digital baseband signal to the required sampling rate, and determine the ratio as the number of downsampling operations;

[0047] Step 132: Convolve the digital baseband signal with the impulse response of the half-band filter, and then perform half-band filtering to extract as one downsampling. Downsample the digital baseband signal according to the number of downsampling operations to obtain the downsampled signal.

[0048] For baseband digital signal r n ' ,f0 Half-band filtering is performed to decimate h[·], thereby converting the sampling rate and obtaining a signal r with a lower sampling rate. n The receiver acquires signals with a high sampling rate. To reduce subsequent computation without affecting the accuracy of sampling decisions, a half-band filter is chosen to achieve a decimation rate that is a power of 2. The impulse response h(k) of the half-band filter is zero at all even points except for zero. Therefore, using a half-band filter to implement sampling rate conversion requires only half the computation, resulting in a high computational efficiency.

[0049]

[0050] Step 2: Divide the downsampled signal continuously according to a fixed window length, and make a decision on the sub-signals within the window to restore the original code.

[0051] Each window contains multiple symbols, each original symbol exists in a symbol period, and each symbol period contains multiple sampling positions;

[0052] When the receiving end receives a symbol, it does not immediately make a decision. Instead, it utilizes the characteristic of continuous phase change in the MSK signal to observe multiple preceding and following signals, improving the accuracy of the symbol decision. An iteration step size of length Q is used, and detection and decision are performed on data with a fixed window length of L each time. Nonlinear filtering is the most effective method for suppressing impulse noise; the nonlinear filter used in this invention includes a decision unit and a limiter.

[0053] Specifically, in one optional embodiment, step 2 includes:

[0054] Step 21: For the downsampled signal r n Perform low-pass filtering f(r) n To filter out out-of-band noise and improve the signal-to-noise ratio of the received signal, the bandwidth of the low-pass filter must be greater than the bandwidth of the target signal.

[0055] Step 22: Divide the low-pass filtered signal continuously according to a fixed window length L to obtain multiple window sub-signals;

[0056] Step 23: Use the decision maker to make a decision on the sub-signals in each window, so that the decision maker is bounded by 0, and obtains the original code elements in each window.

[0057] The decision maker uses 0 as a boundary; if the input sample x is positive, the output y is set to 1; otherwise, it is set to -1. The relationship between the input and output is as follows: Judgment device such as Figure 2 As shown, the decision unit has a simple structure, but its performance is poor because its processing affects the signal.

[0058] Step 3: For each original symbol within a window, take each sampling position within each symbol period as the starting sampling position, and take the window length backward to obtain multiple sampling windows corresponding to each window;

[0059] Specifically, each sampling position within each symbol period is taken as the starting sampling position, and the window length is taken sequentially to obtain multiple sampling window references corresponding to each window. Figure 4 .

[0060] Step 4: Calculate the demodulation signal-to-noise ratio of each sampling window corresponding to each window, select the starting sampling position of the sampling window with the largest demodulation signal-to-noise ratio as the sampling starting point, and obtain the sampling signal in each window by sampling the signal from the sampling starting point;

[0061] The receiver needs to know not only the symbol rate (to sample at the frequency of the symbol rate) but also the exact location within a symbol interval to sample. Typically, timing error changes very slowly relative to the symbol rate; therefore, the timing error over a signal of length L is considered constant. (Reference) Figure 5 In one symbol period T s K different locations within Repeat step 4, taking data with a fixed window length L one at a time. Based on the demodulated signal-to-noise ratio matrix [SNR1, SNR2, ..., SNR...], ... K Select the maximum value SNR of the matrix elements. max The corresponding sampling decision position is taken as the optimal sampling starting point. Sampling is performed at symbol period intervals. In this invention, K is set to 8. The window length L contains M symbols, and one symbol period T... s The window length of L is T. s M; go to k positions within the first symbol.

[0062] Specifically, in one optional embodiment, step 4 includes:

[0063] Step 41: Based on the original symbols of each window, take the positive value of the low-pass filtered sub-signal;

[0064] This invention assigns positive values ​​to the filtered results based on the original symbols of each sampling window. According to the formula... When x is positive, the decision is 1, and the product is positive. When x is negative, the decision is -1, and the product is positive.

[0065] Step 42: Calculate the threshold value of the limiter based on the sub-signal that takes a positive value;

[0066] Step 43: Using this threshold pair, take the positive sub-signals to limit the output amplitude limiting signal;

[0067] Step 43: Calculate the demodulation signal-to-noise ratio of each sampling window corresponding to each window by using the mean and variance of the amplitude-limited signal within the sampling window corresponding to each window.

[0068] Calculate the demodulated signal-to-noise ratio (SNR) of the sub-signal within the calculated window length L. Sudden impulse noise intensity much greater than the signal intensity can cause false start-ups and false stop-ups in detection, degrading detection performance; therefore, this type of noise needs to be suppressed. A limiter, such as... Figure 3 As shown, the threshold of the limiter is crucial to its performance; a reasonable threshold leads to better performance. The limiter is given a threshold λ (λ is a positive constant). When the amplitude of the input signal is within the given range [-λ, λ], the output equals the input; otherwise, the output is either λ or -λ depending on whether the input is positive or negative. The relationship between input and output is shown in the following equation.

[0069]

[0070] To facilitate subsequent processing, this invention first filters the result f(r'). n Take all positive values ​​||f(r') n )||, then f(r') n The signal-to-noise ratio (SNR) is calculated based on the amplitude-limited data after the amplitude-limited value is applied to reduce interference from impulse noise such as lightning. When ||f(r') n When ||≤λ, the output is ||f(r') n If the output is λ, then output λ. Impulse noise does not have an analytical probability density function, but after noise suppression, it can generally be considered as approximately Gaussian noise.

[0071] The threshold value of the limiter in step 42 is

[0072]

[0073] In step 43, the demodulation signal-to-noise ratio for each sampling window is:

[0074]

[0075] Where μ is ||f(r') n The mean of σ|| 2 For ||f(r') n The variance of μ is given by μ. 2 Considered as signal power, σ 2 It can be considered as noise power.

[0076] Step 5: Compare the demodulation signal-to-noise ratio of the sampling window corresponding to the sampled signal with the detection threshold to determine whether the sampled signal is the target signal;

[0077] The demodulated signal-to-noise ratio (SNR) of the sampling window corresponding to the sampled signal. max Compared with the detection threshold γ, when the demodulated signal-to-noise ratio (SNR) is... max If the value is greater than the detection threshold γ, it is confirmed that a target exists within the sampling window, and the sampled signal within the sampling window is taken as the target signal.

[0078] It is worth noting that: the maximum value of the matrix element SNR is selected. max The signal is compared with a set detection threshold γ to achieve the detection of the presence and start / end time of the target signal. When the SNR... max The target signal is considered to exist when the value exceeds the threshold, and from... Initially, sampling points are set for each symbol period; otherwise, the signal is considered nonexistent. If the signal exists, the received signal is traversed to determine the end time of the target signal. The maximum value of the matrix element, SNR, is then used. max If the value is less than the threshold γ, the target signal is considered to have ended; otherwise, the target signal is considered to have not ended.

[0079] Step 6: For the original code corresponding to the target signal in each window and the next window, compare the second half of the original code in the current window with the first half of the original code in the next window to determine whether there is a misalignment in the original code in the next window. If so, correct the original code in the next window to obtain the corrected code of the target signal in each window.

[0080] Considering the continuity of detection and the false negative rate, after the previous set of data is processed, the last data of length M (M < L) should be saved to the beginning of the second set of data. The processing and analysis of each set of data is carried out independently. The drift at the sampling time has little impact in the short term, but as time accumulates, the loss or duplication of sampling symbols will inevitably occur, leading to sequence misalignment, and ultimately the system will be unable to complete the correct demodulation.

[0081] Specifically, in one optional embodiment, step 6 includes:

[0082] Step 61: For the original code elements corresponding to the target signal in each window and the next window, extract the second half of the original code elements in that window. With the first half of the original code in the next window A comparison is performed; if the two do not match, it is determined that the original code elements in the next window have been misaligned.

[0083] Step 62: Determine whether the original code elements in the next window are caused by duplication or missing code elements. If it is caused by duplication, change the first half of the code elements in the next window to... If the problem is caused by missing symbols, then change the first half of the symbols in the next window to...

[0084] refer to Figure 6 as well as Figure 7 As shown, in order to determine if there is a misalignment of the original code elements in the next window, the code elements in the second half of the window are... Perform a left or right shift; if the result of the left shift is similar to the first half of the code in the next window... If they match, then the original code in the next window is determined to be due to duplication; if the result after right shifting matches the first half of the code in the next window... If they match, then it is determined that the original code in the next window was lost.

[0085] To obtain more reliable demodulation results, the last M symbols of the previous data set are taken. The first M symbols of the second set of data Perform a shift comparison, which involves shifting the previous group one position to the front. Shift one position to the right Do not move The final result is the shift with the most identical signs among the three cases, as illustrated in the diagram below. Figure 7 As shown. If the number of identical symbols is at most when shifting forward by one bit, it means that the second group of data is missing a symbol, and it needs to be... Change to The fact that the number of identical symbols is at most when shifted one bit to the right indicates that the second group of data has repeated a symbol, and it needs to be... Change to

[0086] Step 7: Differential decoding is performed on the corrected symbols of the target signal within each window to obtain the decoding result.

[0087] The MSK signal generated by this invention requires differential encoding followed by modulation. Therefore, differential decoding is also required at the receiving end to restore the original digital symbols.

[0088] This invention outputs the start and end times of the target signal and the original symbol, and calculates the bit error rate (BER). BER is an important indicator for analyzing the reliability of a communication system. When analyzing the theoretical performance of the detection algorithm, an MSK signal is generated through simulation and detected via a channel. The BER Pe is calculated according to the following formula:

[0089]

[0090] The effects of this invention can be further illustrated by the following simulation experiments.

[0091] 1. Simulation conditions

[0092] This invention utilizes MATLAB 2021b developed by MathWorks for simulation. The simulation data uses simulated test data, and the signal modulation method is MSK. The simulated received data is obtained by superimposing noise onto the simulated communication signal. MSK is a commonly used signal modulation method in low-frequency communication, featuring continuous phase, constant envelope, and higher spectral efficiency.

[0093] The method compared in the theoretical analysis of the experiment is the traditional energy detection method, which is denoted as ED in the experiment. The reference is N. Kundargi and A. Tewfik, "A performance study of novel Sequential Energy Detection methods for spectrum sensing," 2010 IEEE International Conference on Acoustics, Speech and Signal Processing, 2010, pp. 3090-3093.

[0094] 2. Simulation Content

[0095] (2.a) According to the specific embodiment of the present invention, the signal-to-noise ratio (SNR) of the entire demodulated test data consisting of a 50-second noise signal, a 500-second MSK signal superimposed with noise, and a 50-second noise signal is recorded. The distinguishing effect of the statistical features is observed, and the results are as follows: Figure 8 As shown.

[0096] from Figure 8As can be seen, when the received signal-to-noise ratio (SNR) is low, the signal is completely submerged in noise, making it impossible to effectively distinguish between signal and noise. This invention uses the demodulated SNR as a statistical feature to distinguish between signal and noise, thus achieving signal detection. The demodulated SNR curve shows that this invention effectively improves the value of the statistical feature; the demodulated SNR when a signal is present is significantly higher than when there is only noise. Signal detection can be achieved by selecting an appropriate threshold value, verifying the feasibility of this invention. At low SNR conditions, this invention can directly improve the demodulated SNR value, thereby enhancing detection performance.

[0097] By selecting collected atmospheric data and using software, it can be observed that the target signal persists throughout the sample duration. A 60-second segment is extracted, and 10-second noise is added before and after it to generate the signal to be tested. The method of this invention is then used for detection synchronization to determine the start and end times of the target signal. Figure 9 This is the detection output result. The detection and recognition results match the sample settings.

[0098] (2.b) According to the specific embodiment of the present invention, the detection probability of the signal under different received signal-to-noise ratios is calculated and compared with the detection probability of the ED method, and the result is as follows. Figure 10 As shown, the ED method first calculates the energy of the sample data and then compares it with a threshold value to make a decision. The detection threshold is greatly affected by unknown noise, and it can only calculate the energy of the signal, but cannot distinguish whether it comes from the signal or noise.

[0099] from Figure 10 As can be seen, the present invention achieves better detection results at low signal-to-noise ratios compared to the ED method, further verifying the effectiveness of the present invention. The detection performance at low signal-to-noise ratios is better than that of traditional detection methods.

[0100] (2.c) After simulating and generating the MSK signal and transmitting it through the channel, signal detection and demodulation are performed according to the specific implementation of this invention. The demodulation bit error rate is calculated based on the demodulated bits and transmitted bits, and compared with the theoretical MSK bit error rate. The result is as follows: Figure 11 As shown. The theoretical bit error rate is the MSK coherent demodulation bit error rate.

[0101] from Figure 11 It is evident that the signal-to-noise ratio (SNR) of the received signal has a significant impact on algorithm performance. A higher SNR results in a lower demodulation bit error rate and better algorithm detection performance, while a lower SNR increases the likelihood of signal distortion at the receiving end and leads to a higher bit error rate. Within the range of SNR variation shown in the figure, the simulated bit error rate and the theoretical bit error rate exhibit consistent trends. When Eb / N0 is greater than 8dB, the bit error rate is less than one-thousandth. This invention suppresses impulse noise and reduces timing errors. Although strict carrier synchronization is not performed, it still exhibits good demodulation performance, verifying the reliability of this invention.

[0102] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0103] Although this application has been described herein in conjunction with various embodiments, those skilled in the art will understand and implement other variations of the disclosed embodiments by reviewing the accompanying drawings, the disclosure, and the appended claims in carrying out the claimed application. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude a plurality.

[0104] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A fast signal blind detection and synchronization method under low signal-to-noise ratio, characterized in that, include: Step 1: Receive the wireless analog signal returned from the far field, and perform analog-to-digital conversion, down-conversion, and downsampling on the wireless analog signal to obtain a downsampled signal; Step 2: Divide the downsampled signal continuously according to a fixed window length, and make a decision on the sub-signals within the window to restore the original code. Each window contains multiple symbols, each original symbol exists in a symbol period, and each symbol period contains multiple sampling positions; Step 3: For each original symbol within a window, take each sampling position within each symbol period as the starting sampling position, and take the window length backward to obtain multiple sampling windows corresponding to each window; Step 4: Based on the original symbols of each window, take the positive value of the low-pass filtered sub-signal; based on the sub-signal with the positive value, calculate the threshold value of the limiter; use the threshold value pair to limit the output of the limited signal of the sub-signal with the positive value; use the mean and variance of the limited signal in the sampling window corresponding to each window to calculate the demodulation signal-to-noise ratio of each sampling window corresponding to each window. The threshold value of the limiter is λ (λ is a positive constant). When the amplitude of the input sub-signal is within the threshold value pair [-λ, λ], the output limited signal is equal to the input sub-signal; otherwise, the output limited signal is λ or -λ depending on the sign of the input sub-signal. The relationship between the input sub-signal and the output limited signal is expressed as follows: Select the starting sampling position of the sampling window with the highest demodulation signal-to-noise ratio as the sampling start point, and obtain the sampled signal in each window by sampling the signal from the sampling start point; Step 5: Compare the demodulation signal-to-noise ratio of the sampling window corresponding to the sampled signal with the detection threshold to determine whether the sampled signal is the target signal; Step 6: For the original code corresponding to the target signal in each window and the next window, compare the second half of the original code in the current window with the first half of the original code in the next window to determine whether there is a misalignment in the original code in the next window. If so, correct the original code in the next window to obtain the corrected code of the target signal in each window. Step 7: Differential decoding is performed on the corrected symbols of the target signal within each window to obtain the decoding result.

2. The fast signal blind detection and synchronization method under low signal-to-noise ratio as described in claim 1, characterized in that, Step 1 includes: Step 11: Receive the wireless analog signal returned from the far field and convert the wireless analog signal into a digital signal; Step 12: Down-convert the digital signal to convert it from a digital signal to a digital baseband signal; Step 13: Based on the sampling rate of the digital baseband signal and the required sampling rate, downsample the digital baseband signal to obtain a downsampled signal.

3. The fast signal blind detection and synchronization method under low signal-to-noise ratio as described in claim 2, characterized in that, Step 13 includes: Step 131: Calculate the ratio of the sampling rate of the digital baseband signal to the required sampling rate, and determine the ratio as the number of downsampling operations; Step 132: Convolve the digital baseband signal with the impulse response of the half-band filter, and then perform half-band filtering to extract as one downsampling. Downsample the digital baseband signal according to the number of downsampling operations to obtain the downsampled signal.

4. The fast signal blind detection and synchronization method under low signal-to-noise ratio as described in claim 1, characterized in that, Step 2 includes: Step 21: Perform low-pass filtering on the downsampled signal to remove out-of-band noise; Step 22: Divide the low-pass filtered signal continuously according to a fixed window length L to obtain multiple window sub-signals; Step 23: Use the decision maker to make a decision on the sub-signals in each window, so that the decision maker is bounded by 0, and obtains the original code elements in each window.

5. The fast signal blind detection and synchronization method under low signal-to-noise ratio as described in claim 1, characterized in that, The threshold value of the limiter in step 4 is Where, f(r) n ') represents the sub-signal after low-pass filtering; In step 4, the demodulation signal-to-noise ratio for each sampling window is: Where μ is ||f(r') n The mean of σ|| 2 For ||f(r') n The variance of μ is given by μ. 2 Considered as signal power, σ 2 It can be considered as noise power.

6. The fast signal blind detection and synchronization method under low signal-to-noise ratio as described in claim 1, characterized in that, Step 5 includes: The demodulated signal-to-noise ratio (SNR) of the sampling window corresponding to the sampled signal. max Compared with the detection threshold γ, when the demodulated signal-to-noise ratio (SNR) is... max If the value is greater than the detection threshold γ, it is confirmed that a target exists within the sampling window, and the sampled signal within the sampling window is taken as the target signal.

7. The fast signal blind detection and synchronization method under low signal-to-noise ratio as described in claim 1, characterized in that, Step 6 includes: Step 61: For the original code elements corresponding to the target signal in each window and the next window, extract the second half of the original code elements in that window. With the first half of the original code in the next window A comparison is performed; if the two do not match, it is determined that the original code elements in the next window have been misaligned. Step 62: Determine whether the original code elements in the next window are caused by duplication or missing code elements. If it is caused by duplication, change the first half of the code elements in the next window to... If the problem is caused by missing symbols, then change the first half of the symbols in the next window to...

8. The fast signal blind detection and synchronization method under low signal-to-noise ratio according to claim 7, characterized in that, Step 62 includes: If the original code elements in the next window are misaligned, then the code elements in the first half of the window will be... Perform a left or right shift; if the result of the left shift is similar to the first half of the code in the next window... If they match, then the original code in the next window is determined to be due to duplication; if the result after right shifting matches the first half of the code in the next window... If they match, then it is determined that the original code in the next window was lost.

Images