A low false alarm rate detection method based on spread spectrum sequence signal

Abstract

The application discloses a low false alarm rate detection method based on a spread spectrum sequence signal, and the method comprises the following contents: a direct spread spectrum sequence with a certain length is selected and added to an information sequence as a synchronization signal and a training sequence of underwater acoustic communication; a receiving end firstly performs Doppler multi-channel coarse detection and matched correlation detection on the spread spectrum signal, obtains a corresponding communication signal starting edge through sliding correlation processing, and uses the starting edge to intercept a synchronization signal segment and performs band-pass down-sampling or even under-sampling processing; then, the synchronization signal saved locally is matched and correlated with the obtained signal segment to obtain an estimated channel; after time reversal processing of the estimated channel, secondary detection is performed, and communication signal detection is completed by comparing a correlation peak with a preset threshold. The application can reduce the signal length, reduce the overhead, improve the communication rate, consider the needs of signal synchronization and channel estimation, perform secondary detection to reduce the missed detection and false alarm probability, and ensure the communication quality.

Description

Technical fields:

[0001] This invention belongs to the field of underwater acoustic communication technology, specifically relating to a low false alarm rate detection method based on spread spectrum sequence signals. Background technology:

[0002] As humanity continues to explore the ocean, the demand for underwater information transmission is becoming increasingly urgent. Underwater acoustic communication is a primary communication method capable of long-distance, real-time underwater information exchange, with broad applications not only in the military field but also in civilian applications. However, underwater acoustic communication faces not only the challenge of severely limited bandwidth but also the typical time delay and Doppler spread problems exhibited by underwater acoustic channels under multipath and time-varying conditions.

[0003] Spread spectrum communication (SSC) possesses excellent anti-interference and anti-multipath capabilities, making it the preferred communication method for achieving high-quality underwater acoustic communication. Furthermore, due to the high spread spectrum processing gain at the receiver, SSC underwater acoustic communication systems can operate under low signal-to-noise ratio (SNR) conditions, playing an almost irreplaceable role in long-range / ultra-long-range underwater acoustic communication. The correct despreading and decoding of SSC underwater acoustic communication generally involves key processes such as signal detection and synchronization, channel estimation, and equalization. Among these, signal detection and synchronization are prerequisites for subsequent high-quality decoding / decoding. SSC underwater acoustic communication has strong anti-interference and noise capabilities, so it is generally chosen in low SNR environments to improve the robustness and detection probability of the communication system. However, in low SNR underwater acoustic environments, SSC communication systems face a significant problem of high false alarm rates in signal detection. When a large number of false signals exist during real-time signal detection, it consumes excessive computing resources. Moreover, the communication system platform has limited resources, leading to missed detections when genuine signals arrive, resulting in a decline in system performance.

[0004] Common underwater acoustic communication signal frame structures typically include modules such as synchronization signals, signal gaps, channel estimation sequences, and information sequences. Traditional spread-spectrum underwater acoustic communication detection methods detect LFM (linear frequency modulation) and HFM (hyperbolic frequency modulation) signals, which serve as synchronization sequences. However, in low signal-to-noise ratio environments, the matching gain drops significantly. Spread-spectrum sequences are frequently used in channel estimation due to their excellent autocorrelation and cross-correlation properties. However, due to the complexity of underwater acoustic channels, bandwidth limitations, strong low-frequency noise interference, and real-time requirements in practical applications, spread-spectrum underwater acoustic communication has a relatively low communication rate, making the interception and detection of spread-spectrum underwater acoustic signals more difficult and limiting its practical applications. Summary of the Invention:

[0005] The technical problem this invention aims to solve is to address the issue of numerous false signals during signal detection in spread spectrum underwater acoustic communication under low signal-to-noise ratio and high interference conditions, leading to missed detection of real signals and a high false alarm rate. This invention provides a low false alarm rate detection method based on spread spectrum sequence signals. This method can reduce the occurrence of false alarms in low signal-to-noise ratio environments, improve the detection probability of spread spectrum signals, and exhibits good detection performance. It is suitable for spread spectrum underwater acoustic communication, can significantly improve the reliability of remote communication systems, and can reduce the redundancy of underwater acoustic communication signal frame structures, eliminating the overhead of synchronization headers, thus possessing excellent practicality.

[0006] The technical solution of this invention is to provide a low false alarm rate detection method based on spread spectrum sequence signals, which includes the following:

[0007] A direct spread spectrum sequence of a certain length is selected and added to the information sequence as the synchronization signal and training sequence for underwater acoustic communication. The receiver first performs Doppler multi-channel coarse detection and matched correlation detection on the spread spectrum signal, and obtains the corresponding communication signal start edge through sliding correlation processing. The synchronization signal segment is truncated using the start edge and bandpass downsampling or even undersampling processing is performed. Then, the locally stored synchronization signal is matched correlation processed with the obtained signal segment to obtain the estimated channel. After time inversion processing using the estimated channel, a second detection is performed. The communication signal detection is completed by comparing the correlation peak with a preset threshold.

[0008] Spread spectrum sequences have good time-width-bandwidth product and correlation performance. Therefore, this invention proposes a low false alarm rate signal detection method based on pseudo-random spread spectrum sequences for signal synchronization and channel estimation. It does not require frequency modulation sequences and uses the same spread spectrum sequence to simultaneously perform communication signal frame synchronization and channel estimation. This reduces signal length, reduces overhead, and increases communication rate, while also meeting the needs of signal synchronization and channel estimation. Furthermore, it performs secondary detection to reduce the probability of missed detections and false alarms, thus ensuring communication quality.

[0009] In this system, a spread spectrum sequence of a certain length can serve as both a synchronization signal for effective signal detection and an accurate estimation of the underwater acoustic channel, achieving the effect of multiplexing the synchronization signal and channel estimation. The spread spectrum sequence has excellent matching gain, resulting in good signal detection performance. In low signal-to-noise ratio (SNR) and strong interference environments, it avoids missed detections, lowers the detection threshold, and improves detection efficiency. Furthermore, its relatively accurate channel estimation allows for time-reversal processing, obtaining additional channel focusing gain and enhancing the system's anti-interference capability. Moreover, to adapt to low SNR and strong interference environments, underwater acoustic communication based on spread spectrum sequences lowers the detection threshold. First-level threshold detection of the spread spectrum sequence reduces missed detections, and subsequent time-reversal processing of the estimated channel followed by second-level threshold detection effectively solves the problem of distinguishing between spread spectrum and interference signals, significantly reducing the system's false alarm rate.

[0010] Preferably, the method includes the following key steps:

[0011] (1) After the information sequence has completed channel coding, interleaving and other operations, the transmitter adds the spread spectrum sequence before the information sequence or superimposes it with the information sequence, and sends it out after carrier modulation as a synchronization signal and channel estimation sequence.

[0012] (2) Signal synchronization: Signal detection and synchronization are performed at the receiving end. The detection method involves directly performing matched correlation processing between the received signal and multiple locally stored copies of the Doppler coarse interpolated spread spectrum sequence. To reduce computational load, the received signal is downsampled at equal intervals before the matched correlation processing. If the ratio of the synchronization signal's center frequency to its bandwidth is large, bandpass undersampling technology can be used to reduce the number of sample points. The correlation peak value of the received signal and the starting edge of the synchronization signal are obtained through sliding correlation processing of the copied signal. The starting edge plus the length of the synchronization signal is the position of the end of the synchronization signal, and the synchronization signal of the corresponding length is truncated.

[0013] (3) Channel estimation: The received signal in step (1) is matched and correlated with the local spread spectrum sequence. This will match multiple obvious correlation peaks and numerous small peaks. To further optimize the calculation, we set the small peaks below the highest peak to zero to obtain the estimated value of the underwater acoustic channel at the current moment.

[0014] (4) Time-reversal processing: The estimated underwater acoustic channel and the received signal are time-reversed to improve the spread spectrum gain.

[0015] (5) Secondary detection: After obtaining the received signal after time-reversal processing in step (4), repeat step (1) to perform secondary matching correlation processing on the received signal to obtain the correlation peak value. The detection of the synchronization signal is completed by comparing this average value with the preset threshold size.

[0016] Compared with the prior art, the present invention has the following advantages:

[0017] This invention eliminates the need for a frequency modulation (FM) sequence, using the same spread spectrum sequence for simultaneous communication signal frame synchronization and channel estimation. This reduces signal length, overhead, and communication rate while simultaneously meeting the requirements for signal synchronization and channel estimation. Furthermore, secondary detection reduces the probability of missed detections and false alarms, ensuring communication quality. Specifically, the transmitting end uses a spread spectrum sequence of a certain length as both the synchronization sequence and the channel estimation sequence, reducing the overhead of using the original FM signal as the synchronization sequence. The channel estimated by the spread spectrum sequence is used to perform time-reversal processing on the received signal before secondary detection, effectively reducing missed detections and false peak misjudgments. Especially under low signal-to-noise ratio (SNR) conditions, lowering the detection threshold significantly improves detection performance. This method exhibits strong anti-interference capabilities, is suitable for complex underwater acoustic communications, and can significantly enhance the robustness of long-distance communication. Attached image description:

[0018] Figure 1 This is a schematic diagram of the workflow of the spread spectrum sequence synchronization and channel estimation multiplexing detection method.

[0019] Figure 2 The diagram shows the time-inverse matching correlation results before and after processing of a 1023-length spread spectrum sequence under typical channel conditions. In the diagram, 2(a) is the result of time-inverse preprocessing, and 2(b) is the result of time-inverse postprocessing.

[0020] Figure 3 This is a diagram showing the channel estimation results for a 1023-length spread spectrum sequence under different signal-to-noise ratios.

[0021] Figure 4 This is a graph showing the results of inverse pre- and post-matching correlation processing for a 1023-length spread spectrum sequence under different signal-to-noise ratios.

[0022] Figure 5 This is a graph showing the detection-false alarm probability curves of the detection method of this invention and conventional detection methods under different detection threshold conditions. Detailed implementation method:

[0023] The present invention will be further described below with reference to the accompanying drawings and specific embodiments:

[0024] A low false alarm rate detection method based on spread spectrum sequence signals is illustrated in the flowchart below. Figure 1 As shown, the specific implementation process of this method is as follows:

[0025] (1) The transmitter encodes and interleaves the original information to form an information data sequence. A pseudo-random spreading sequence of a certain length M, c(t) = [c1, c2, ..., c...], is added before the information sequence. M As a training sequence, each spreading chip has a width of T. i Chip rate R i =1 / T i ,but For information codes, As a gate function, M = 1023 is used in this example. After modulation, a single frame of communication signal is finally formed:

[0026]

[0027] The received signal model is:

[0028]

[0029] In the formula: y(t) is the received signal, w(t) is the additive noise, J(t) is the interference, and A is the amplitude of the received signal.

[0030] The despreading process at the receiver is the same as the spreading process. The received signal is demodulated after correlation calculation using the locally stored spreading sequence. The user obtains the spreading gain, and noise interference and other interference components are significantly weakened due to energy dispersion after despreading. Let the locally stored spreading sequence be c'(t), then the despreading process can be expressed by the formula:

[0031]

[0032] In the formula, the signal, noise and interference after despreading are denoted as s'(t), w'(t) and J'(t), respectively.

[0033] If the spread spectrum sequence generated by the received signal is synchronized with the local spread spectrum sequence, then c'(t) = c(t), that is, c'(t)c(t) = 1, and the receiver will get a relatively obvious correlation peak A.

[0034] (2) Signals are received in real time during signal detection. The communication signal containing the spread spectrum sequence is the signal to be detected. A binary hypothesis detection model is used to describe the signal sensing problem in underwater acoustic communication:

[0035]

[0036] Where y(n) is the signal to be detected received by the receiver, s(n) is the speech signal containing the HFM guidance sequence, and w(n) is the additive white Gaussian noise in the measured signal, which conforms to w(n)~N h(n) and k represent the channel and the number of sampling points, respectively; H0 indicates that there is no signal to be detected in the current received signal; H1 indicates that the signal to be detected has been detected and the next step of signal analysis and processing can be carried out.

[0037] When detecting signals, use f * (n) and g * (n) represents the energy normalization of the transmitted signal f(n) and the received signal g(n), R1(t) represents the correlation between the locally copied signal and the signal to be detected, and ρ1 represents the normalized correlation value:

[0038]

[0039]

[0040] ρ1=max(R1(n)) (7)

[0041] Multiple Doppler coarse interpolation local copy signals are subjected to sliding correlation processing with the signal to be detected, which can improve the anti-Doppler capability of the spread spectrum signal. A larger correlation value is selected and compared with the selection threshold γ1. When the communication channel noise interference is large and the signal-to-noise ratio is low, the matched correlation peak is small. In order to adapt to the low signal-to-noise ratio environment, reduce missed detections, and improve the resistance to noise and interference, the threshold γ1 is set to a small value. When ρ1≥γ1, it means that the secondary spread spectrum sequence signal to be detected is detected. The starting edge of the corresponding spread spectrum signal is obtained by using the position of the correlation peak. The starting edge plus the length of the spread spectrum synchronization signal is the position of the end of the synchronization signal. The spread spectrum signal of the corresponding length is truncated.

[0042] (3) The received signal from step (2) is subjected to matching correlation processing with the local spread spectrum sequence, such as... Figure 2 As shown in (a), multiple obvious correlation peaks and numerous smaller peaks will be matched. To further optimize the calculation, we can artificially reduce or set to zero a certain proportion of the smaller peaks that are lower than the highest peak to reduce the interference of pseudo-peaks and obtain the estimated value h'(n) of the underwater acoustic channel at the current moment. It has the basic characteristic information of the current channel. The spread spectrum sequence performs well in channel estimation. Figure 3 The results are channel estimation results for a spread spectrum sequence of length M=1023 under a certain signal-to-noise ratio. It can be seen that under low signal-to-noise ratio (-10dB to -8dB), in addition to estimating the original channel, there are some additional spurious peaks, which increase the interference. When the signal-to-noise ratio is greater than -7dB, the main peak of the estimated channel is obvious, and there are basically no spurious peaks, indicating that the spread spectrum sequence has good channel estimation performance.

[0043] (4) The time-reversal mirror focuses the underwater acoustic multipath channel in time and space to reduce multipath signal interference. Taking the virtual time-reversal mirror as an example, the estimated channel h'(n) is time-reversed, and then the obtained time-reversed channel h'(-n) is convolved with the received signal y(n) to realize the entire time-reversal process.

[0044] After performing channel estimation and time-reversal processing on the secondary signal to be detected, a second matched correlation processing is performed to obtain the time-reversal processing gain. From formulas (3) and (4), it can be seen that the received signal after time reversal and despreading is:

[0045]

[0046] In the formula: A' is the relevant amplitude, The time-inverse channel can be considered as the channel through which the signal ultimately passes. It can be expressed as the cross-correlation function of the actual channel and the estimated channel. Since the estimated channel h'(n) and the real channel h(n) have similar basic characteristics, the time-inverse channel can be considered as the autocorrelation function of the actual channel and the estimated channel, achieving the superposition of multipath signal energy and producing a focusing effect. Figure 2(b) is the result of time-inversion processing under typical channel conditions, and... Figure 2 (a) Improved matching correlation processing gain compared to the previous method. Figure 4 These are the correlation peak values ​​of the received signal after time-inversion processing and without time-inversion processing under different signal-to-noise ratio conditions. It can be clearly seen that time inversion greatly increases the matching processing gain, has a good energy focusing effect, especially under low signal-to-noise ratio conditions, and has strong anti-interference ability.

[0047] Next, the locally copied signal and the time-inverted signal are subjected to matched correlation processing and compared with the threshold γ2. Due to the time-inverted processing gain, the peak value of the second matched correlation will be significantly increased. When the second correlation value ρ2 ≥ γ2, it indicates that a signal has been detected. If it is not a legitimate spread spectrum signal, the second correlation value ρ2 will not exceed the selection threshold γ2. The total false alarm probability can be expressed as:

[0048]

[0049] in: and Let represent the false alarm probabilities of a single correlation detection and a second correlation detection after time-reversal processing, respectively. By using a two-level detection threshold to reduce the probability of false alarms, the signal can be further processed. Figure 5 The detection-false alarm probability curves under different detection thresholds are presented. The results show that the method of the present invention can significantly reduce the false alarm probability while improving the detection probability compared with conventional detection methods, and the signal detection performance is better.

[0050] Typically, when the signal bandwidth is fixed, a higher sampling rate means the signal bandwidth and energy remain constant, but the noise introduced into the signal widens, causing fluctuations in the product value within segments. Furthermore, a higher sampling rate also increases the computational load. To maintain consistent overall system performance and reduce computational load, the received signal can be downsampled at equal intervals. If the ratio of the signal center frequency to the bandwidth is large, bandpass undersampling technology can also be used to reduce the number of sampling points and related computational load.

[0051] The detection method proposed in this invention uses the same spreading sequence at the transmitting end as both a synchronization sequence and a channel estimation sequence. This reduces the overhead of using the original FM signal as a synchronization sequence. The excellent autocorrelation properties of the spreading sequence further improve the accuracy of channel estimation. The channel estimated by the spreading sequence is then used to perform time-reversal processing on the received signal for secondary detection. This significantly reduces the impact of noise and interference on the signal, improves the gain of matched correlation processing, and effectively reduces detection omissions and false peak misjudgments, especially under low signal-to-noise ratio conditions, thus significantly improving detection performance. This method has strong anti-interference capabilities and is suitable for complex underwater acoustic communication, particularly significantly improving the robustness of long-distance communication.

[0052] The above description only illustrates preferred embodiments of the present invention and should not be construed as limiting the scope of the claims. Any equivalent procedural modifications made using this specification are included within the patent protection scope of this invention.

Claims

1. A low false alarm rate detection method based on spread spectrum sequence signals, characterized in that: The method includes, A direct spread spectrum sequence of predetermined length is added to the information sequence as the synchronization signal and training sequence for underwater acoustic communication. The receiver first performs Doppler multi-channel coarse detection and matched correlation detection on the spread spectrum signal, and obtains the corresponding communication signal start edge through sliding correlation processing. The synchronization signal segment is truncated using the start edge, and bandpass downsampling or even undersampling processing is performed. Then, the locally stored synchronization signal is matched correlation processed with the obtained signal segment to obtain the estimated channel. After time-inverse processing using the estimated channel, a second detection is performed. The communication signal detection is completed by comparing the correlation peak with a preset threshold. The specific steps are as follows: Step 1: After the transmitter completes channel coding and interleaving correlation operations on the information sequence, it adds the spread spectrum sequence before the information sequence or superimposes it on the information sequence, which serves as both a synchronization signal and a channel estimation sequence. After carrier modulation, the spread spectrum sequence is transmitted. Step 2, signal synchronization: Signal detection and synchronization processing are performed at the receiving end. The detection method is to directly match the received signal with multiple copies of the spread spectrum sequence after coarse Doppler interpolation stored locally. The corresponding correlation peak and the start edge of the synchronization signal are obtained by copying the signal and sliding correlation processing. The start edge plus the length of the synchronization signal is the position of the end of the synchronization signal. The synchronization signal of the corresponding length is then truncated. Step 3, channel estimation: set the smaller peaks that are below a certain percentage of the highest peak to zero to obtain the estimated value of the underwater acoustic channel at the current moment; Step four, time-reversal processing: The estimated underwater acoustic channel and the received signal are time-reversed to improve the spreading gain. Step 5, secondary detection: After obtaining the received signal after time-reversal processing in Step 4, repeat Step 1 to perform secondary matching correlation processing on the received signal to obtain the correlation peak value. The detection of the synchronization signal is completed by comparing this average value with the preset threshold value.

2. The low false alarm rate detection method based on spread spectrum sequence signals according to claim 1, characterized in that: In step two, to reduce the computational load, the received signal is downsampled at equal intervals before matching correlation processing. If the ratio of the center frequency to the bandwidth of the synchronization signal reaches a specific value, bandpass undersampling technology is further adopted to reduce the number of sample points.

3. The low false alarm rate detection method based on spread spectrum sequence signals according to claim 1, characterized in that: In step two, signals are received in real time during signal detection. The communication signal containing the spread spectrum sequence is the signal to be detected. A binary hypothesis detection model is used to describe the signal sensing problem in underwater acoustic communication. in, The signal to be detected received by the receiving end. It is a speech signal containing an HFM guidance sequence; It is additive white Gaussian noise in the measured signal, and conforms to... ; and These represent the number of channels and the number of sampling points, respectively. This indicates that there is no signal to be detected in the currently received signal; This indicates that the signal to be tested has been detected.

4. The low false alarm rate detection method based on spread spectrum sequence signals according to claim 3, characterized in that: During signal detection, respectively using and Indicates the transmitted signal and received signals Energy normalization, using This indicates the correlation between the locally copied signal and the signal to be detected. Represents the normalized correlation values: Multiple local copy signals after coarse Doppler interpolation are subjected to sliding correlation processing with the signal under test to improve the anti-Doppler capability of the spread spectrum signal.

5. The low false alarm rate detection method based on spread spectrum sequence signals according to claim 4, characterized in that: when The time indicates that a secondary spread spectrum sequence signal to be detected has been detected. The starting edge of the corresponding spread spectrum signal is obtained by using the position of the correlation peak. The starting edge plus the length of the spread spectrum synchronization signal is the position of the end of the synchronization signal. The spread spectrum signal of the corresponding length is then extracted.

Images