阅读说明：本技术 一种主动声纳单频脉冲串波形设计及检测算法 (Active sonar single-frequency pulse train waveform design and detection algorithm ) 是由 岳雷 于 2019-10-23 设计创作，主要内容包括：本发明公开一种主动声纳单频脉冲串波形设计及检测算法,首先是根据探测需求及硬件条件等设计波形参数。设计好参数后,一方面通过发射机发射信号,将接收回波数据正交解调降采样后经过自动增益控制,然后再通过多普勒滤波器组；另一方面设置副本信号,将副本信号正交解调降采样后通过多普勒滤波器组。最后将经过多普勒滤波器组后的两段数据分段相关处理得到一个输出。该方法方法计算量小,实时性强,采用了波形分集技术,一定程度上解决了混响对主动声纳探测的影响,改善了主动声纳探测的效果,具有较好的工程应用前景。(The invention discloses an active sonar single-frequency pulse train waveform design and detection algorithm. After the parameters are designed, on one hand, the signals are transmitted through a transmitter, the received echo data are subjected to orthogonal demodulation and down sampling, then are subjected to automatic gain control, and then pass through a Doppler filter bank; and on the other hand, a replica signal is set, and the replica signal is subjected to quadrature demodulation and down sampling and then passes through a Doppler filter bank. And finally, performing segmentation correlation processing on the two sections of data after passing through the Doppler filter bank to obtain an output. The method has the advantages of small calculated amount and strong real-time performance, adopts the waveform diversity technology, solves the influence of reverberation on active sonar detection to a certain extent, improves the effect of the active sonar detection, and has better engineering application prospect.)

1. An active sonar single-frequency pulse train waveform design and detection algorithm is characterized in that: comprises the following steps of (a) carrying out,

step S1: designing single-frequency pulse signal parameters, wherein the single-frequency pulse signal parameters are designed to have a center frequency of 5kHz, an effective duration of 10ms of sub-pulses, a duty ratio of 0.4, the number of the sub-pulses of 4 and a total duration of 100ms, wherein 5 comb teeth are arranged in an effective bandwidth of a parameter frequency spectrum, and then transmitting signals through a transmitter and receiving echo data;

step S2: orthogonal demodulation and down sampling, orthogonal demodulation is respectively carried out on the transmitting signal and the echo, the calculation formula is as follows,

s_c(t)＝exp[2s(t)exp(j2πf_dt)]_LPF(1)

r_c(t)＝exp[2r(t)exp(j2πf_dt)]_LPF(2)

step S3: setting replica signals, setting a scale factor and the number of replicas according to a target speed range to be detected and a speed tolerance of the signals, wherein the set M replica signals are respectively

s_m(t)＝s(η_mt) (3)

Wherein M is 1,2,. M;

step S4: is provided with a plurality ofThe filter set comprises comb filters arranged on the frequency axis of the single-frequency pulse signal, with comb width of 2/T and comb interval of 1/T_pObtaining the Comb filter Comb _ f with the scale factor of 1₁(ii) a From Comb _ f_m＝Comb_f₁·η_mCalculating comb filters corresponding to other scales, and combining the filters corresponding to all the scales to form an M-channel comb filter group;

step S5: target detection and speed and distance parameter estimation, firstly, the received echo data is self-correlated, then zero-filling is carried out, then Fourier transform is carried out to obtain a section of output data X_mThen X is added_mRespectively accumulating and summing after passing through M channels of the filter bank to obtain M output values, and finding out the maximum value and a channel n corresponding to the maximum value to obtain the target speed v_nThe calculation formula is as follows:

Out_m＝∑X_m(j)·Comb_f_m(j) (4)

Out_n＝max([Out₁,Out₂,……Out_M]) (5)

n＝find(Out_n＝＝[Out₁,Out₂,……Out_M]) (6)

then, a matched replica signal s is constructed according to the estimated target speed_nAfter zero padding, Fourier transform is carried out to obtain S_nObtaining R after Fourier transform after zero filling of received echo data_n' the calculation formula is as follows,

and finally, selecting the maximum value of the test statistics as a final test statistic st, and finishing the estimation of the target time delay at the same time, wherein the calculation formula is as follows,

st＝max([st₁,st₂,……st_K]) (8)

if the test statistic is larger than the preset threshold, the target exists, otherwise, the target does not exist, and the process is repeated to detect the next section of data.

2. The active sonar single-frequency burst waveform design and detection algorithm of claim 1, wherein: the single-frequency pulse signal has the following calculation formula,

3. the active sonar single-frequency burst waveform design and detection algorithm of claim 1, wherein: the number of copies is not less than the ratio of the target range to the signal speed tolerance.

4. The active sonar single-frequency burst waveform design and detection algorithm of claim 1, wherein: the frequency spectrum of the single-frequency pulse signal is

Wherein P (f) is the Fourier transform of a single FM sub-pulse signal and has

5. The active sonar single-frequency burst waveform design and detection algorithm of claim 3, wherein: the width of the main lobe and the grating lobe is 2/T, and the distance between the main lobe and the grating lobe is 1/T_pSo that the single-frequency pulse signal spectrum has comb teeth width of 2/T and comb teeth interval of 1/T_pThe comb spectrum of (1).

Technical Field

The invention belongs to the field of underwater active sonar detection, and particularly relates to an active sonar single-frequency pulse train waveform design and detection algorithm.

Background

The existing active sonar generally adopts a single-frequency signal or a frequency modulation signal to perform underwater sound detection. For white gaussian noise background, the matched filter is the optimal detector, and the maximum output signal-to-noise ratio can be obtained as long as the pulse width of the transmitted signal is large enough. However, the underwater acoustic environment is more and more complex and cannot be a white gaussian noise background, the underwater acoustic detection is difficult due to the occurrence of reverberation, and how to effectively suppress the reverberation is a problem to be solved urgently in need of the current active sonar. The long single-frequency pulse signal has better anti-reverberation performance on a high-speed moving target, and the frequency modulation signal has better anti-reverberation performance on a static target. In order to further improve the anti-reverberation performance of the active sonar, a new waveform design and detection algorithm needs to be considered.

Disclosure of Invention

In order to solve the problems and the defects in the prior art, the invention provides an active sonar single-frequency pulse train waveform design and detection algorithm.

The following design structure and design scheme are specifically adopted:

an active sonar single-frequency pulse train waveform design and detection algorithm comprises the following steps:

step S1: designing single-frequency pulse signal parameters, wherein the single-frequency pulse signal parameters are designed to have a center frequency of 5kHz, an effective duration of 10ms of sub-pulses, a duty ratio of 0.4, the number of the sub-pulses of 4 and a total duration of 100ms, wherein 5 comb teeth are arranged in an effective bandwidth of a parameter frequency spectrum, and then transmitting signals through a transmitter and receiving echo data;

step S2: orthogonal demodulation and down sampling, orthogonal demodulation is respectively carried out on the transmitting signal and the echo, the calculation formula is as follows,

s_c(t)＝exp[2s(t)exp(j2πf_dt)]_LPF(1)

r_c(t)＝exp[2r(t)exp(j2πf_dt)]_LPF(2)

step S3: setting replica signals, setting a scale factor and the number of replicas according to a target speed range to be detected and a speed tolerance of the signals, wherein the set M replica signals are respectively

s_m(t)＝s(η_mt) (3)

Wherein M is 1,2,. M;

step S4: setting Doppler filter group, setting comb filter on the frequency axis of single-frequency pulse signal, taking 2/T as comb width and 1/T as comb interval_pObtaining the Comb filter Comb _ f with the scale factor of 1₁(ii) a From Comb _ f_m＝Comb_f₁·η_mCalculating comb filters corresponding to other scales, and combining the filters corresponding to all the scales to form an M-channel comb filter group;

step S5: target detection and speed and distance parameter estimation, firstly, the received echo data is self-correlated, then zero-filling is carried out, then Fourier transform is carried out to obtain a section of output data X_mThen X is added_mRespectively accumulating and summing after passing through M channels of the filter bank to obtain M output values, and finding out the maximum value and a channel n corresponding to the maximum value to obtain the target speed v_nThe calculation formula is as follows:

Out_m＝∑X_m(j)·Comb_f_m(j) (4)

Out_n＝max([Out₁,Out₂,……Out_M]) (5)

n＝find(Out_n＝＝[Out₁,Out₂,……Out_M]) (6)

then, a matched replica signal s is constructed according to the estimated target speed_nAfter zero padding, Fourier transform is carried out to obtain S_nObtaining R after Fourier transform after zero filling of received echo data_n' the calculation formula is as follows,

and finally, selecting the maximum value of the test statistics as a final test statistic st, and finishing the estimation of the target time delay at the same time, wherein the calculation formula is as follows,

st＝max([st₁,st₂,……st_K]) (8)

if the test statistic is larger than the preset threshold, the target exists, otherwise, the target does not exist, and the process is repeated to detect the next section of data.

Preferably, the single-frequency pulse signal has the following calculation formula,

preferably, the number of copies is not less than the ratio of the target range to the signal speed margin.

Preferably, the frequency spectrum of the single-frequency pulse signal is

Wherein P (f) is the Fourier transform of a single FM sub-pulse signal and has

Preferably, the | Y (f) | refers to the width of the main lobe and the grating lobe being 2/T, and the distance between the main lobe and the grating lobe being 1/T_pSo that the single-frequency pulse signal spectrum has comb teeth width of 2/T and comb teeth interval of 1/T_pThe comb spectrum of (1).

The working principle of the invention is as follows:

aiming at the influence of underwater reverberation interference on the detection performance of the active sonar, the invention designs a waveform with better reverberation resistance on a moving target in a certain speed range by using the spectral characteristics of a single-frequency pulse train signal, then sets a Doppler filter bank to filter part of reverberation according to the spectral characteristics of the waveform to improve the signal-to-noise ratio, and finally adopts a correlator to finish detection.

Compared with the prior art, the invention has the following beneficial effects: the method has small calculated amount and strong real-time performance, adopts the waveform diversity technology, solves the influence of reverberation on active sonar detection to a certain extent, improves the effect of the active sonar detection, and improves the detection performance of static and moving targets.

Drawings

FIG. 1 is a schematic flow chart of the present invention.

Fig. 2 is a time domain spectrum diagram of a single-frequency pulse signal.

FIG. 3 is a graph of the Q function of the signal.

FIG. 4 is a graph showing the detection output at a speed of 0 m/s.

FIG. 5 is a graph showing the detection output at a speed of 1 m/s.

FIG. 6 is a graph showing the detection output at a speed of 2 m/s.

FIG. 7 is a graph showing the detection output at a speed of 3 m/s.

FIG. 8 is a graph showing the detection output at a speed of 4 m/s.

FIG. 9 is a graph showing the detection output at a speed of 5 m/s.

Detailed Description

The following describes embodiments of the present invention in more detail with reference to the accompanying drawings and specific examples.

Example 1: as shown in fig. 1 of the specification, an active sonar single-frequency pulse train waveform design and detection algorithm includes the following steps:

step S1: designing single-frequency pulse signal parameters, wherein the single-frequency pulse signal parameters are designed to have a center frequency of 5kHz, an effective duration of 10ms of sub-pulses, a duty ratio of 0.4, the number of the sub-pulses of 4 and a total duration of 100ms, wherein 5 comb teeth are arranged in an effective bandwidth of a parameter frequency spectrum, and then transmitting signals through a transmitter and receiving echo data;

step S2: orthogonal demodulation and down sampling, orthogonal demodulation is respectively carried out on the transmitting signal and the echo, the calculation formula is as follows,

s_c(t)＝exp[2s(t)exp(j2πf_dt)]_LPF(1)

r_c(t)＝exp[2r(t)exp(j2πf_dt)]_LPF(2)

step S3: setting replica signals, setting a scale factor and the number of replicas according to a target speed range to be detected and a speed tolerance of the signals, wherein the number of replicas is not less than a ratio of the target range to the speed tolerance of the signals, and the set M replica signals are respectively

s_m(t)＝s(η_mt) (3)

Wherein M is 1,2,. M;

step S4: setting a Doppler filter bank, the frequency spectrum of the single-frequency pulse signal is

Wherein P (f) is the Fourier transform of a single FM sub-pulse signal and has

The width of the main lobe and the grating lobe is 2/T, and the distance between the main lobe and the grating lobe is 1/T_pSo that the single-frequency pulse signal spectrum has comb teeth width of 2/T and comb teeth interval of 1/T_pThe comb spectrum of (2) sets up a Doppler filter bank according to the spectrum characteristics of the multi-Prevoter combined single-frequency pulse signal of the target to be detected, and specifically comprises the following steps: a comb filter is arranged on the frequency axis of the single-frequency pulse signal, the width of the comb teeth is 2/T, and the interval of the comb teeth is 1/T_pObtaining the Comb filter Comb _ f with the scale factor of 1₁(ii) a From Comb _ f_m＝Comb_f₁·η_mCalculating comb filters corresponding to other scales, and combining the filters corresponding to all the scales to form an M-channel comb filter group;

step S5: target detection and speed and distance parameter estimation, firstly, the received echo data is self-correlated, then zero-filling is carried out, then Fourier transform is carried out to obtain a section of output data X_mThen X is added_mRespectively accumulating and summing after passing through M channels of the filter bank to obtain M output values, and finding out the maximum value and a channel n corresponding to the maximum value to obtain the target speed v_nThe calculation formula is as follows:

Out_m＝∑X_m(j)·Comb_f_m(j) (4)

Out_n＝max([Out₁,Out₂,……Out_M]) (5)

n＝find(Out_n＝＝[Out₁,Out₂,……Out_M]) (6)

then, a matched replica signal s is constructed according to the estimated target speed_nAfter zero padding, Fourier transform is carried out to obtain S_nAfter zero filling, the received echo data is Fourier transformedAfter substitution R is obtained_n' the calculation formula is as follows,

and finally, selecting the maximum value of the test statistics as a final test statistic st, and finishing the estimation of the target time delay at the same time, wherein the calculation formula is as follows,

st＝max([st₁,st₂,……st_K]) (8)

if the test statistic is larger than the preset threshold, the target exists, otherwise, the target does not exist, and the process is repeated to detect the next section of data.

Example 2: as shown in the attached fig. 2-3 of the specification, the active sonar single-frequency pulse train waveform is designed as follows: the single-frequency pulse signal has the following calculation formula,

in shallow sea environments, reverberation is the dominant disturbance of active sonar. The anti-reverberation waveform design and the corresponding signal processing algorithm are basic approaches to solve the target detection in the reverberation background. Taking a Q function as a tool for analyzing the reverberation resistance of the signal, wherein the smaller the value of the Q function is, the better the reverberation resistance is shown, wherein the expression of the Q function is as follows:

the Q function of the short pulse CW signal is basically consistent in a larger speed range, the Q function of the long pulse CW signal is reduced along with the increase of the speed, the Q function of the single-frequency pulse train signal is reduced and then increased along with the change of the speed, and the reciprocating motion is in periodic change. It can be seen from the figure that the Q function value of the mono-frequency burst signal is the smallest when the speed is 1.1m/s to 3.6m/s, that is, the anti-reverberation performance of the mono-frequency burst signal is the best in this speed range.

Example 3: the waveform design and detection algorithm verification of the single-frequency pulse signal are completed in a computer simulation mode, and a submarine reverberation model is established. In simulation, the target is assumed to be located at 1.23s, and the target is in a background of reverberation and noise interference, wherein the signal-to-noise ratio is 0dB and the mixing-to-noise ratio is 0 dB. The simulation was run for 100 times and if the target was found at 1.2/1.25s, the target was deemed to be detected correctly. The detection output graphs are shown in FIGS. 4-9 at target speeds of 0m/s, 1m/s, 2m/s, 3m/s, 4m/s, and 5m/s, respectively.

The scope of the present invention is not limited to the above-described embodiments, which are intended to help explain and illustrate the present invention, but not to limit the scope of the present invention, if it is designed to be the same as or substituted by the equivalent design of the present invention, and fall within the scope of the present invention as claimed.