Weak target robust high-gain detection method using continuous wave MIMO sonar and Doppler filtering

Abstract

In order to solve the problems of short weak target detection distance, long detection time, few detection cycles and the like of the existing traditional active sonar, the invention provides a weak target robust high-gain detection method by using continuous wave MIMO sonar and Doppler filtering. According to the method, frequency diversity continuous wave pulse signals with low cross correlation are designed according to the number of transmitting transducers and the number of detection periods of the MIMO sonar, then the designed frequency diversity continuous wave pulse signals are continuously transmitted by using the MIMO sonar and a multi-period transmitting mode, then Doppler filtering processing is performed on echoes, and the frequency diversity continuous wave pulse signals with low cross correlation are obtained. Doppler filtering output of all frequency diversity continuous wave pulse signals is obtained, coherent accumulation and non-coherent accumulation are carried out on the Doppler filtering output, stable high-gain accumulation output is obtained, in the process, Doppler filtering of the continuous wave pulse signals can be fully utilized to suppress reverberation, multi-period accumulation is utilized to suppress noise, and the noise suppression effect is improved. And the weak target detection performance superior to that of the traditional active sonar is obtained.

Description

Technical Field

[0001] The invention relates to the field of array signal processing, and in particular to a weak target robust high-gain detection method using continuous wave MIMO sonar and Doppler filtering. Background Art

[0002] With the continuous improvement of active stealth capabilities of underwater acoustic targets, the target intensity has been significantly reduced, and the range of active sonar has also been significantly reduced. This is because when the target is at a long distance, the target echo becomes very weak, and the active sonar is faced with low signal-to-mixing ratio and low signal-to-noise ratio working conditions. For the above reasons, how to use active sonar to detect targets of interest at a longer distance has become a hot topic and difficulty in current research.

[0003] At present, in the technical field related to active sonar detection, the main active sonar system used is a single-element transmission and multiple-element reception (SIMO: single-input multiple-output). This type of traditional active sonar mainly uses a repeated periodic detection working mode during detection. In order to avoid mutual interference between echoes between adjacent cycles, the traditional active sonar needs to set a sufficiently long detection cycle interval to ensure that the scattered signals of distant targets are all returned before continuing to transmit the detection signal of the next cycle. Although the use of a sufficiently long detection cycle can improve the target detection capability, due to the long detection cycle interval, non-cooperative targets can confront, resulting in too few cycles that the traditional active sonar can accumulate in a limited time, resulting in a low multi-cycle cumulative gain, making it difficult to effectively detect weak targets in a short time, and ultimately detection failure.

[0004] Some scholars use continuous wave active sonar (CAS) to transmit high duty cycle signals to obtain higher data update rates, thereby improving the detection capability of active sonar. However, the number of detection cycles obtained by the existing CAS sonar using a single transmission signal is limited, and a long detection time is required, which is difficult to meet the needs of weak target detection. In order to obtain multiple detection cycles in a short time and improve the potential of weak target detection performance, multiple-input multiple-output (MIMO) sonar can also be used. However, the number of signals emitted by the MIMO sonar in the existing technology in a single cycle is limited by the number of transmitting transducers, which is difficult to meet the needs of weak target detection.

[0005] In summary, the existing target detection methods using active sonar are limited by defects such as short detection distance, long detection time, and small number of detection cycles. Therefore, we urgently need to propose a method with weak target detection performance that is superior to traditional active sonar. Summary of the invention

[0006] In order to overcome the problems of short weak target detection distance, long detection time and small number of detection cycles of existing traditional active sonar, the present invention proposes a robust high-gain detection method for weak targets using continuous wave MIMO sonar and Doppler filtering, which can fully utilize the Doppler filtering of continuous wave pulse signals to suppress reverberation and utilize multi-cycle accumulation to suppress noise, thereby obtaining weak target detection performance that is superior to traditional active sonar.

[0007] The technical solution of the invention is:

[0008] A robust high-gain detection method for weak targets using continuous wave MIMO sonar and Doppler filtering is characterized by comprising the following steps:

[0009] Step 1, designing a frequency diversity continuous wave pulse signal with low cross-correlation according to the number of transmitting transducers and the number of detection cycles of the MIMO sonar;

[0010] Step 2, continuously transmitting the designed frequency diversity continuous wave pulse signal using MIMO sonar and multi-cycle transmission mode;

[0011] Step 3: Doppler filtering is performed on the received waves to obtain Doppler filtering outputs of all frequency diversity continuous wave pulse signals;

[0012] Step 4: Perform coherent accumulation and incoherent accumulation on the Doppler filter output to obtain a robust and high-gain cumulative output.

[0013] Furthermore, the step 1 specifically includes:

[0014] First, a single-base MIMO sonar is constructed: M transmitting transducers are selected to form a transmitting array, where M is an integer; N hydrophones are selected to form a multi-element receiving array, where the hydrophones are arranged in a straight line, where N is an integer; the axis direction of the multi-element receiving array is set as the x direction, and the direction perpendicular to the x direction is set as the y direction, so that the transmitting array and the receiving array together form a MIMO sonar;

[0015] Redesign the transmitting waveform: the number of transmitting transducers is designed to be M, the number of detection cycles is K, then the number of frequency diversity continuous wave pulse signals is the product of the number of transmitting transducers and the number of detection cycles, which is MK; the frequency diversity continuous wave pulse signals have the same pulse width T and the same frequency interval Δf, where the frequency interval Δf needs to be large enough to ensure that the peak value of the cross-correlation function is not greater than 0.001 times the peak value of the autocorrelation function; at the same time, it is necessary to design a sufficiently large frequency interval Δf based on the maximum Doppler frequency shift generated by the target movement speed to ensure that the Doppler frequency shift of the target echo will not cause echo confusion of adjacent frequency diversity continuous wave pulse signals.

[0016] Furthermore, in step 2:

[0017] M transmitting array elements transmit M frequency diversity continuous wave pulse signals simultaneously. The frequency diversity continuous wave pulse signal emitted by the mth transmitting transducer in the kth detection cycle is expressed as:

[0018]

[0019] In the formula,

[0020] s m,k (t) represents the pulse signal emitted by the mth transmitting transducer in the kth detection cycle, where m = 1, 2, ..., M, k = 1, 2, ..., K;

[0021] f m,k represents the center frequency of the pulse signal of the mth transmitting transducer in the kth detection cycle;

[0022] t represents time;

[0023] T represents the pulse width of the transmitting signal;

[0024] ΔT represents the time interval between adjacent detection cycles, and the value of ΔT is much smaller than the time interval used by traditional active sonar, that is, the value range of ΔT is ΔT∈(0, 6T];

[0025] A represents the signal amplitude, and Where F represents the sampling frequency;

[0026] j represents the imaginary number symbol.

[0027] Furthermore, the step 3 specifically includes:

[0028] Step 3.1: The MIMO sonar simultaneously transmits M mutually orthogonal continuous pulse signals in each detection cycle, which form target echoes after irradiating the target. The multi-element receiving array receives the target echoes, and performs time domain beamforming on the received echoes to obtain time domain beam output;

[0029] Step 3.2: Set a possible range of radial velocity values ​​of the target, discretize the radial velocity of the target within the velocity range, and obtain L discrete radial velocity values;

[0030] Step 3.3: Combine the parameters of the transmitted MK frequency diversity continuous wave pulse signals and design a matched filter at each discrete speed, so that each frequency diversity continuous wave pulse signal can obtain a matched filter group consisting of L matched filters;

[0031] Step 3.4: Design MK matched filter groups for the transmitted MK frequency diversity continuous wave pulse signals, wherein each group of matched filters has L matched filters in total, and use the MK matched filter groups to process the echo signals to obtain the results of MK detection cycles;

[0032] Step 3.5: The output results of the L matched filters are combined to obtain the processing result of the matched filter group designed for the pulse signal emitted by the mth transmitting transducer in the kth detection period on the received echo.

[0033] Furthermore,

[0034] The processing result of the matched filter group designed for the pulse signal emitted by the mth transmitting transducer in the kth detection period in step 3.5 on the received echo is:

[0035] Y m,k (t,v)=[y m,k,1 (t,v) y m,k,2 (t,v) … y m,k,L (t,v)]

[0036] In the formula,

[0037] Y m,k (t,v) represents the processing result of the mth and kth matched filter groups on the received waveform;

[0038] y m,k,l The matched filter output obtained by processing the received signal by the lth matched filter of the matched filter group designed for the pulse signal emitted by the mth transmitting transducer in the kth detection period is expressed as:

[0039]

[0040] In the formula,

[0041] Indicates the received echo corresponding to MK transmitted pulse signals;

[0042] * indicates convolution operation;

[0043] h m,k,l (t) = s m,k,l c (Tt) represents the impulse response function of the lth matched filter designed with the parameters of the pulse signal emitted by the mth transmitting transducer in the kth detection period, wherein the superscript c represents conjugation, and the subscripts m, k, l represent the sequence number corresponding to the pulse signal emitted by the mth transmitting transducer in the kth detection period and the lth discrete velocity value;

[0044] τ m,k =(k-1)(ΔT+T)+2R / c, k=1,2,…K represents the time delay of the signal emitted by the mth transmitting transducer in the kth detection cycle, where R represents the distance between the target and the sonar;

[0045] v represents the radial velocity of the target; c represents the speed of sound; r(t) represents reverberation; n(t) represents noise;

[0046] represents the Doppler frequency shift signal designed by the pulse signal emitted by the mth transmitting transducer in the kth detection cycle and the lth discrete velocity value, where v l represents the lth discrete radial velocity.

[0047] Furthermore, the step 4 specifically includes:

[0048] The complex domain outputs of all MK matched filter banks are summed to obtain the coherent accumulation output of the continuous wave MIMO sonar, which is expressed as:

[0049]

[0050] In the formula,

[0051] Y C (t,v) represents the coherent accumulation output of the continuous wave MIMO sonar;

[0052] The absolute value outputs of all MK matched filter groups are summed to obtain the incoherent cumulative output of the continuous wave MIMO sonar, which is expressed as:

[0053]

[0054] In the formula,

[0055] Y InC (t,v) represents the incoherent accumulation output of the continuous wave MIMO sonar.

[0056] The beneficial effects of the present invention are:

[0057] In view of the problem that traditional active sonar mainly adopts a repeated periodic detection working mode during detection and needs to set a sufficiently long detection cycle, which makes it difficult to accumulate too few cycles within a limited time and thus makes it difficult to detect weak targets, the present invention proposes a frequency diversity continuous wave pulse signal designed for continuous transmission using MIMO sonar and a multi-cycle transmission mode, combined with Doppler filtering, incoherent accumulation, coherent accumulation and other methods, fully utilizing the Doppler filtering of the continuous wave pulse signal to suppress reverberation, utilizing multi-cycle accumulation to suppress noise, and obtaining a robust and high-gain cumulative output.

[0058] The basic principle of the present invention has been theoretically derived, and the implementation plan has been verified by computer numerical simulation. The weak target detection results obtained by the traditional method and the method proposed by the present invention are respectively given, and the results show that the method proposed by the present invention can be applied to various MIMO sonar arrays, and the signals used can be designed according to detection requirements, such as detecting weak targets, detecting long-distance targets, detecting underwater slow small targets and other application scenarios; reverberation and noise can be suppressed at the same time to obtain weak target detection performance that is superior to traditional active sonar; in practical applications, different numbers of transmitting array elements and detection cycles can also be designed according to requirements to achieve robust high-gain detection of weak targets by active sonar. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The above and / or other features and advantages of the present invention will become more easily understood through the following description with reference to the accompanying drawings, which are not drawn to scale and some features are exaggerated or reduced to show details of specific components. In the accompanying drawings:

[0060] Figure 1 Flowcharts of the main steps of the present invention, wherein (a) is a flow chart of the main steps of the coherent accumulation process, and (b) is a flow chart of the main steps of the incoherent accumulation process;

[0061] Figure 2 It is a schematic diagram of the transmitting mode of the transmitting transducer of the present invention;

[0062] Figure 3 Schematic diagrams of array structures of MIMO sonars, wherein (a) is a schematic diagram of the array structure of the MIMO sonar in implementation example simulation 1, (b) is a schematic diagram of the array structure of the MIMO sonar in implementation example simulation 2, (c) is a schematic diagram of the array structure of the MIMO sonar in implementation example simulation 3, (d), (e), and (f) are examples of array structures that can be used in the present invention;

[0063] Figure 4 The spectrum of the transmitted signal used by the MIMO sonar in the implementation example simulation 1, where (a) is a total of 10 FDCW signals transmitted, (b) is the transmitted signal of the No. 1 transmitting transducer, and (c) is the transmitted signal of the No. 2 transmitting transducer;

[0064] Figure 5 The autocorrelation function and the highest cross-correlation function of the transmitted signal used by the MIMO sonar in the implementation example simulation 1;

[0065] Figure 6 Detection results of implementation example simulation 1, where (a) is a single-cycle detection result, (b) is a detection result of coherent accumulation processing, and (c) is a detection result of incoherent accumulation processing;

[0066] Figure 7The transmission signal spectrum used for the MIMO sonar in the implementation example simulation 2, where (a) is a total of 12 FDCW signals transmitted, (b) is the transmission signal of the No. 1 transmitting transducer, (c) is the transmission signal of the No. 2 transmitting transducer, and (d) is the transmission signal of the No. 3 transmitting transducer;

[0067] Figure 8 The autocorrelation function and the highest cross-correlation function of the transmitted signal used for the MIMO sonar in the implementation example simulation 2;

[0068] Fig. 9 Detection results of implementation example simulation 2, where (a) is a single-cycle detection result, (b) is a detection result of coherent accumulation processing, and (c) is a detection result of incoherent accumulation processing;

[0069] Fig.10 The spectrum of the transmitted signal used in the MIMO sonar in the implementation example simulation 3, where (a) is a total of 10 FDCW signals transmitted, and (b) is a time-frequency diagram of the transmitted signal, where the FDCW pulse signal is transmitted simultaneously by 5 transmitting transducers;

[0070] Fig.11 The autocorrelation function and the highest cross-correlation function of the transmitted signal used for the MIMO sonar in the implementation example simulation 3;

[0071] Fig.12 Detection results of implementation example simulation 3, where (a) is a single-cycle detection result, (b) is a detection result of coherent accumulation processing, and (c) is a detection result of incoherent accumulation processing. DETAILED DESCRIPTION

[0072] The present invention will be described in detail below with reference to the accompanying drawings. It should be noted that the following detailed description of the present invention is only for the purpose of illustration rather than limitation of the present invention.

[0073] The present invention proposes a robust high-gain detection method for weak targets using continuous wave MIMO sonar and Doppler filtering, such as Figure 1 FIG. 1 is a flowchart of the main steps of the present invention, including two types of coherent accumulation processing and incoherent accumulation processing, wherein (a) is a flowchart of the main steps of the coherent accumulation processing, and (b) is a flowchart of the main steps of the incoherent accumulation processing. Figure 2 The figure shows a schematic diagram of the transmitting mode of the transmitting transducer of the present invention. Each step of the present invention is described in detail below:

[0074] Step 1: Design a frequency diverse continuous wave (FDCW) pulse signal with low cross-correlation according to the number of transmitting transducers and the number of detection cycles of the MIMO sonar.

[0075] First, a single-base MIMO sonar is constructed: M transmitting transducers are selected to form a transmitting array, where M is an integer; N hydrophones are selected to form a multi-element receiving array, where the hydrophones are arranged in a straight line, where N is an integer; the axial direction of the multi-element receiving array is set as the x direction, and the direction perpendicular to the x direction is set as the y direction, then the transmitting array and the receiving array together constitute a MIMO sonar.

[0076] Redesign the transmitting waveform: the number of transmitting transducers is designed to be M, the number of detection cycles is designed to be K, then the number of FDCW pulse signals is the product of the number of transmitting transducers and the number of detection cycles, which is MK; the FDCW pulse signals have the same pulse width T and the same frequency interval Δf, where the frequency interval Δf needs to be large enough to ensure that the peak value of the cross-correlation function is not greater than 0.001 times the peak value of the autocorrelation function; at the same time, it is necessary to design a sufficiently large frequency interval Δf based on the maximum Doppler frequency shift generated by the target movement speed to ensure that the Doppler frequency shift of the target echo will not cause echo confusion of adjacent FDCW pulse signals.

[0077] Step 2: Use MIMO sonar and multi-cycle transmission mode to continuously transmit the designed FDCW pulse signal.

[0078] The transmitting array simultaneously transmits M mutually orthogonal continuous wave pulse signals. Taking the FDCW pulse signal as an example, M transmitting array elements simultaneously transmit M FDCW pulse signals. By analogy, all MK FDCW pulse signals are transmitted in K detection cycles. Then, the pulse signal expression emitted by the mth transmitting transducer in the kth detection cycle is:

[0079]

[0080] In the formula,

[0081] s m,k (t) represents the pulse signal emitted by the mth transmitting transducer in the kth detection cycle, where m = 1, 2, ..., M, k = 1, 2, ..., K;

[0082] f m,k represents the center frequency of the pulse signal of the mth transmitting transducer in the kth detection cycle;

[0083] t represents time;

[0084] T represents the pulse width of the transmitting signal;

[0085] ΔT represents the time interval between adjacent detection cycles, and the value of ΔT is much smaller than the time interval used by traditional active sonar, that is, the value range of ΔT is ΔT∈(0, 6T];

[0086] A represents the signal amplitude, and Where F represents the sampling frequency;

[0087] j represents the imaginary number symbol.

[0088] Step 3: Perform Doppler filtering on the received waves to obtain the Doppler filtering output of all FDCW pulse signals.

[0089] Step 3.1: The MIMO sonar simultaneously transmits M mutually orthogonal continuous pulse signals in each detection cycle, which form target echoes after irradiating the target. The multi-element receiving array receives the target echoes, and performs time domain beamforming on the received echoes to obtain time domain beam output.

[0090] Assuming that the target is located in a certain beam and the distance between the target and the center of the sonar system is R, the echo on the beam can be expressed as the superposition of multiple continuous wave pulses with different parameters (center frequency, delay) after passing through the channel, that is, the expression of the received echo corresponding to all MK transmitted pulse signals transmitted in K detection cycles is:

[0091]

[0092] In the formula,

[0093] x(t) represents the received echo corresponding to MK transmitted pulse signals;

[0094] τ m,k =(k-1)(ΔT+T)+2R / c, k=1,2,…K represents the time delay of the signal emitted by the mth transmitting transducer in the kth detection cycle;

[0095] v represents the radial motion speed of the target;

[0096] c represents the speed of sound;

[0097] r(t) represents reverberation;

[0098] n(t) represents noise.

[0099] Step 3.2: Assume that the radial velocity v of the target (unless otherwise specified, the velocity in the following text refers to the radial velocity of the relative motion between the target and the sonar) can take a value range of v∈[v rd1 ,v rd2 ], where v rd1 , v rd2 represents the radial velocity of the target. The radial velocity of the target is discretized within the above velocity range to obtain L discrete radial velocity values. The lth discrete radial velocity v l The expression is:

[0100]

[0101] Step 3.3: Based on the parameters of the transmitted MK FDCW pulse signals, a set of matched filters is designed for each pulse signal, that is, a matched filter is designed at each discrete speed, and each FDCW pulse signal can obtain a matched filter group consisting of L matched filters.

[0102] The expression of the impulse response function of the lth matched filter designed with the parameters of the pulse signal emitted by the mth transmitting transducer in the kth detection cycle is:

[0103] h m,k,l (t) = s m,k,l c (Tt) (4)

[0104] In the formula,

[0105] h m,k,l (t) represents the impulse response function of the lth matched filter designed with the parameters of the pulse signal emitted by the mth transmitting transducer in the kth detection period;

[0106] The superscript c represents conjugation, and the subscripts m, k, and l represent the serial numbers corresponding to the pulse signal emitted by the mth transmitting transducer in the kth detection cycle and the lth discrete velocity value;

[0107] s m,k,l (t) represents the Doppler frequency shift signal designed by the pulse signal emitted by the mth transmitting transducer in the kth detection cycle and the lth discrete velocity value, and its expression is:

[0108]

[0109] Step 3.4: Design MK matched filter groups for the transmitted MK FDCW pulse signals, where each group of matched filters has L matched filters, as expressed by equations (4) and (5). Use MK matched filter groups to process the echo signals to obtain the results of MK detection cycles. The corresponding matched filter output is:

[0110] y m,k,l =x(t)*h m,k,l (t) (6)

[0111] In the formula,

[0112] y m,k,l represents the matched filter output obtained by processing the received signal by the lth matched filter of the matched filter group designed for the pulse signal transmitted by the mth transmitting transducer in the kth detection period;

[0113] * indicates a convolution operation.

[0114] Since the transmission signal is designed according to the maximum Doppler frequency shift generated by the target speed, the transmission signal has a sufficiently large frequency interval Δf to ensure that the Doppler frequency shift of the target echo will not cause echo confusion of adjacent FDCW pulse signals. Therefore, the matched filter output obtained by processing the received signal by the lth matched filter of the matched filter group designed by the pulse signal transmitted by the mth transmitting transducer in the kth detection cycle can be expressed as:

[0115]

[0116] Step 3.5: Combining the output results of L matched filters can obtain the processing result of the matched filter group designed for the pulse signal emitted by the mth transmitting transducer in the kth detection cycle to the received echo:

[0117] Y m,k (t,v)=[y m,k,1 (t,v) y m,k,2 (t,v) … y m,k,L (t,v)] (8)

[0118] In the formula,

[0119] Y m,k (t,v) represents the processing result of the m,k-th matched filter groups on the received waveform.

[0120] The MK group matched filter outputs, i.e., the results of MK detection cycles, are obtained by using the MK group matched filters for subsequent processing.

[0121] Step 4: Perform coherent accumulation and incoherent accumulation on the Doppler filter output to obtain a robust and high-gain cumulative output.

[0122] The present invention uses coherent accumulation processing and incoherent accumulation processing to fuse the results of MK detection cycles to obtain a weak target detection result that is better than that of a single detection cycle, wherein:

[0123] The complex domain outputs of all MK matched filter banks are summed to obtain the coherent accumulation output of the continuous wave MIMO sonar, which is expressed as:

[0124]

[0125] In the formula,

[0126] Y C (t,v) represents the coherent accumulation output of the continuous wave MIMO sonar.

[0127] The absolute value outputs of all MK matched filter groups are summed to obtain the incoherent cumulative output of the continuous wave MIMO sonar, which is expressed as:

[0128]

[0129] In the formula,

[0130] Y InC (t,v) represents the incoherent accumulation output of the continuous wave MIMO sonar.

[0131] Implementation Examples

[0132] The present invention will be described in detail below with the aid of an exemplary implementation example of the present invention, that is, numerical simulation is performed using a computer to verify the effect of the method proposed in the present invention.

[0133] Simulation 1: M = 2, K = 5

[0134] The number of transmitting array elements of MIMO sonar is M=2, the center of the transmitting array element coordinate is located at the origin, the number of receiving array ULA elements is N=8, and the array structure diagram is as follows Figure 3 As shown in (a), the number of transmitting transducers is set to M=2, the number of detection cycles is K=5, the sampling frequency is 10kHz, the frequency interval of adjacent FDCW signals is Δf=50Hz, the pulse width of a single pulse is T=2 seconds, and the signal amplitude is A=0.0071.

[0135] Perform Fourier transform on the transmitted signal, and the obtained transmitted signal is a combination of 10 FDCW signals, and its spectrum is as follows: Figure 4 As shown; calculate the autocorrelation function and cross-correlation function of the transmitted signal, and the results are as follows Figure 5 As shown, the peak height of the cross-correlation function is about 50 dB lower than the peak height of the autocorrelation function.

[0136] Assume that there is a target in the detection scene, the distance between the target and the sonar is R = 20km, the radial speed of the target is v = 3m / s, 100 scattering points are set to simulate reverberation echo, the signal-to-mixing ratio is about -20dB, and complex Gaussian white noise is added to each frame reception, and its power is 10dB. The processing results of the single frame reception echo are as follows: Figure 6 As shown in (a), the results of coherent accumulation and incoherent accumulation of 10 frames of received echo are as follows Figure 6 (b) Figure 6 (c) as shown.

[0137] It can be seen from the detection results that the traditional single-cycle detection has a high background and is difficult to detect the target, while the coherent accumulation and incoherent accumulation processing can reduce the background and detect the target with a radial motion speed of v = 3m / s at 20km.

[0138] Simulation 2: M = 3, K = 4

[0139] The number of transmitting array elements of MIMO sonar is M=3, the center of the transmitting array element coordinate is located at the origin, the number of receiving array ULA elements is N=8, and the array structure diagram is as follows Figure 3 As shown in (b), the number of transmitting transducers is set to M=3, the number of detection cycles is set to K=4, and the other parameters are the same as those in simulation 1.

[0140] Performing Fourier transform on the transmitted signal, the obtained transmitted signal is a combination of 12 FDCW signals, and its spectrum is as follows: Figure 7 As shown; calculate the autocorrelation function and cross-correlation function of the transmitted signal, and the results are as follows Figure 8 As shown, the peak height of the cross-correlation function is about 50 dB lower than the peak height of the autocorrelation function.

[0141] Assume that there is a target in the detection scene, the distance between the target and the sonar is R = 20km, the radial speed of the target is v = 3m / s, 100 scattering points are set to simulate reverberation echo, the signal-to-mixing ratio is about -20dB, and complex Gaussian white noise is added to each frame reception, and its power is 13dB. The processing results of the single frame reception echo are as follows: Fig. 9 As shown in (a), the results of coherent accumulation and incoherent accumulation of 12 frames of received echo are as follows Fig. 9 (b) Fig. 9 (c) as shown.

[0142] It can be seen from the detection results that due to the insufficient reverberation and noise suppression capabilities, the traditional single-cycle detection has a high background and is difficult to detect the target. The coherent accumulation and incoherent accumulation processing proposed by the method of the present invention can detect moving targets with a distance of 20km and a radial motion speed of v=3m / s.

[0143] Simulation 1 and simulation 2 show that the proposed method can obtain better reverberation and noise suppression capabilities than traditional methods, thereby effectively improving the weak target detection performance of active sonar. In practical applications, different numbers of transmitting array elements and detection cycles can be designed according to requirements to achieve robust high-gain detection of weak targets by active sonar.

[0144] Simulation 3: M=5, K=2 slow small target detection

[0145] The number of transmitting array elements of MIMO sonar is M=5, the center of the transmitting array element coordinates is located at the origin, the number of receiving array ULA elements is N=8, and the array structure diagram is as follows Figure 3 As shown in (c), the number of transmitting transducers is set to M=5, the number of detection cycles is K=2, the sampling frequency is 100kHz, the frequency interval Δf between adjacent FDCW signals is 500Hz, the signal frequency band is 30.5kHz-35kHz, the pulse width of a single pulse is T=0.2 seconds, and the signal amplitude is A=0.0071.

[0146] Perform Fourier transform on the transmitted signal, and the obtained transmitted signal is a combination of 10 FDCW signals, and its spectrum is as follows: Fig.10 As shown. Calculate the autocorrelation function and cross-correlation function of the transmitted signal, the results are as follows Fig.11 As shown, the peak height of the cross-correlation function is about 50 dB lower than the peak height of the autocorrelation function.

[0147] Assume that there is a slow small target in the detection scene, the distance between the target and the sonar is R = 2km, the radial speed of the target is v = 1m / s, 100 scattering points are set to simulate reverberation echo, the signal-to-mixing ratio is about -20dB, and complex Gaussian white noise is added to each frame reception, and its power is 10dB. The processing results of the single frame reception echo are as follows: Fig.12 As shown in (a), the results of coherent accumulation and incoherent accumulation of 10 frames of received echo are as follows Fig.12 (b) Fig.12 (c) as shown.

[0148] It can be seen from the detection results that the traditional single-cycle detection has a high background and is difficult to detect the target, while the coherent accumulation and incoherent accumulation processing can reduce the background and detect the target with a radial motion speed of v = 1m / s at 2km. Therefore, the method proposed in the present invention can also be used for slow small target detection.

[0149] According to the implementation examples, it can be considered that the method proposed in the present invention can be applied to various MIMO sonar arrays, and the signals used can be designed according to the detection requirements, such as detecting weak targets, detecting long-distance targets, detecting slow small underwater targets, etc. The present invention uses a method for detecting weak targets using continuous wave MIMO sonar, Doppler filtering, and multi-cycle accumulation, which effectively solves the problem that traditional active sonar is difficult to effectively detect weak targets.

Claims

1. A robust high-gain detection method for weak targets using continuous wave MIMO sonar and Doppler filtering, characterized in that: The following steps are involved: Step 1, designing a frequency diversity continuous wave pulse signal with low cross-correlation according to the number of transmitting transducers and the number of detection cycles of the MIMO sonar; Step 2, continuously transmitting the designed frequency diversity continuous wave pulse signal using MIMO sonar and multi-cycle transmission mode; Step 3: Doppler filtering is performed on the received waves to obtain Doppler filtering outputs of all frequency diversity continuous wave pulse signals; Step 4: Perform coherent accumulation and incoherent accumulation on the Doppler filter output to obtain a robust and high-gain cumulative output.

2. The method for robust high-gain detection of weak targets using continuous wave MIMO sonar and Doppler filtering as claimed in claim 1, characterized in that: The step 1 specifically includes: First, a single-base MIMO sonar is constructed: M transmitting transducers are selected to form a transmitting array, where M is an integer; N hydrophones are selected to form a multi-element receiving array, where the hydrophones are arranged in a straight line, where N is an integer; the axis direction of the multi-element receiving array is set as the x direction, and the direction perpendicular to the x direction is set as the y direction, so that the transmitting array and the receiving array together form a MIMO sonar; Redesign the transmitting waveform: the number of transmitting transducers is designed to be M, the number of detection cycles is K, then the number of frequency diversity continuous wave pulse signals is the product of the number of transmitting transducers and the number of detection cycles, which is MK; the frequency diversity continuous wave pulse signals have the same pulse width T and the same frequency interval Δf, where the frequency interval Δf needs to be large enough to ensure that the peak value of the cross-correlation function is not greater than 0.001 times the peak value of the autocorrelation function; at the same time, it is necessary to design a sufficiently large frequency interval Δf based on the maximum Doppler frequency shift generated by the target movement speed to ensure that the Doppler frequency shift of the target echo will not cause echo confusion of adjacent frequency diversity continuous wave pulse signals.

3. The method for robust high-gain detection of weak targets using continuous wave MIMO sonar and Doppler filtering as claimed in claim 1, characterized in that: In step 2: M transmitting array elements transmit M frequency diversity continuous wave pulse signals simultaneously. The frequency diversity continuous wave pulse signal emitted by the mth transmitting transducer in the kth detection cycle is expressed as: In the formula, s m,k (t) represents the pulse signal emitted by the mth transmitting transducer in the kth detection cycle, where m = 1, 2, ..., M, k = 1, 2, ..., K; f m,k represents the center frequency of the pulse signal of the mth transmitting transducer in the kth detection cycle; t represents time; T represents the pulse width of the transmitting signal; ΔT represents the time interval between adjacent detection cycles, and the value of ΔT is much smaller than the time interval used by traditional active sonar, that is, the value range of ΔT is ΔT∈(0, 6T]; A represents the signal amplitude, and Where F represents the sampling frequency; j represents the imaginary number symbol.

4. The method for robust high-gain detection of weak targets using continuous wave MIMO sonar and Doppler filtering as claimed in claim 1, characterized in that: The step 3 specifically includes: Step 3.1: The MIMO sonar simultaneously transmits M mutually orthogonal continuous pulse signals in each detection cycle, which form target echoes after irradiating the target. The multi-element receiving array receives the target echoes, and performs time domain beamforming on the received echoes to obtain time domain beam output; Step 3.2: Set a possible range of radial velocity values ​​of the target, discretize the radial velocity of the target within the velocity range, and obtain L discrete radial velocity values; Step 3.3: Combine the parameters of the transmitted MK frequency diversity continuous wave pulse signals and design a matched filter at each discrete speed, so that each frequency diversity continuous wave pulse signal can obtain a matched filter group consisting of L matched filters; Step 3.4: Design MK matched filter groups for the transmitted MK frequency diversity continuous wave pulse signals, wherein each group of matched filters has L matched filters in total, and use the MK matched filter groups to process the echo signals to obtain the results of MK detection cycles; Step 3.5: The output results of the L matched filters are combined to obtain the processing result of the matched filter group designed for the pulse signal emitted by the mth transmitting transducer in the kth detection period on the received echo.

5. The method for robust high-gain detection of weak targets using continuous wave MIMO sonar and Doppler filtering as claimed in claim 4, characterized in that: The processing result of the matched filter group designed for the pulse signal emitted by the mth transmitting transducer in the kth detection period in step 3.5 on the received echo is: Y m,k (t,v)=[y m,k,1 (t,v) y m,k,2 (t,v)…y m,k,L (t,v)] In the formula, Y m,k (t,v) represents the processing result of the mth and kth matched filter groups on the received echo; y m,k,l The matched filter output obtained by processing the received signal by the lth matched filter of the matched filter group designed for the pulse signal emitted by the mth transmitting transducer in the kth detection period is expressed as: In the formula, Indicates the received echo corresponding to MK transmitted pulse signals; * indicates convolution operation; h m,k,l (t) = s m,k,l c (Tt) represents the impulse response function of the lth matched filter designed with the parameters of the pulse signal emitted by the mth transmitting transducer in the kth detection period, wherein the superscript c represents conjugation, and the subscripts m, k, l represent the sequence number corresponding to the pulse signal emitted by the mth transmitting transducer in the kth detection period and the lth discrete velocity value; τ m,k =(k-1)(ΔT+T)+2R / c, k=1,2,…K represents the time delay of the signal emitted by the mth transmitting transducer in the kth detection cycle, where R represents the distance between the target and the sonar; v represents the radial velocity of the target; c represents the speed of sound; r(t) represents reverberation; n(t) represents noise; represents the Doppler frequency shift signal designed by the pulse signal emitted by the mth transmitting transducer in the kth detection cycle and the lth discrete velocity value, where v l represents the lth discrete radial velocity.

6. The method for robust high-gain detection of weak targets using continuous wave MIMO sonar and Doppler filtering as claimed in claim 1, characterized in that: The step 4 specifically includes: The complex domain outputs of all MK matched filter banks are summed to obtain the coherent accumulation output of the continuous wave MIMO sonar, which is expressed as: In the formula, Y C (t,v) represents the coherent accumulation output of the continuous wave MIMO sonar; The absolute value outputs of all MK matched filter groups are summed to obtain the incoherent cumulative output of the continuous wave MIMO sonar, which is expressed as: In the formula, Y InC (t,v) represents the incoherent accumulation output of the continuous wave MIMO sonar.

Images