| Sonar detection systems and relevant technologies have been greatly developed.However,sonar systems still face significant challenges such as the gradual reducing intensity of target radiated noise and the gradual reducing target intensity.To obtain the same spatial resolution as the high-frequency sonars,the sonars require a larger array aperture when the desired signal frequency becomes lower.Due to various restrictions,it is not suitable for some small platforms to equip large aperture sonar arrays.An acoustic vector sensor or a small aperture acoustic vector sensors array can provide a better choice.The acoustic vector sensor is small in size,light in weight and convenient to deploy,making it suitable for installation on some small platforms.The passive detection is generally difficult to accurately estimate the detected target position,especially using an acoustic vector sensor or a small aperture acoustic vector sensor array.In order to determine the target position,an active sonar system is employed for detection and localization.The distributed sonar system based on acoustic vector sensor can effectively extend the detection coverage and improve the performance of the equipment system,so it has broad application prospects.This paper focuses on target detection problem for a small platform in the active and the passive sonar system using the acoustic vector sensor as the receiver.Several techniques are proposed to improve the detection and localization performance of the active and the passive sonar systems.This paper focuses on and aims to solve the following issues:1.The output characteristics of sound pressure and vibration velocity time reversal convolution joint processing and its application in signal processing.Two convolution joint processing methods using sound pressure and vibration velocity are studied.The first proposed processing output for the line spectrum signals includes the signal autocorrelation function,the noise autocorrelation function,and the cross term noise,whilst that of the other proposed method includes signal autocorrelation function and cross term noise.Convolution joint processing can obtain spatial gain and suppress noise,thereby improving signal detection performance.Firstly,a convolution joint processing method is proposed to process the difference between signal and noise output,then improve the performance of line spectrum signal detection by post adaptive enhancement in a low SNR condition.In addition,another convolution joint processing method can be used to improve the detection method of wideband signals.The simulation and experimental data processing results show that the performance of line spectrum detection can be improved by the convolution joint processing using the post adaptive enhancement.2.Azimuth estimation of line spectrum signal based on convolution joint processing outputs using adaptive cancellation.Aiming at the shortcomings of small directivity gain and low accuracy of azimuth estimation for sound pressure and vibration velocity joint processing using an acoustic vector sensor,an azimuth estimation method based on the adaptive cancellation output of convolution joint processing of sound pressure and vibration velocity is proposed.According to the output characteristics of the convolution joint processing method,the adaptive canceller eliminates only the input signal noise in the target orientation,whilst eliminates the line spectrum autocorrelation function and noise in the non-target orientation.Then,the target azimuth is estimated according to the residual spatial distribution curve of the adaptive canceller.Compared with other methods,the simulation results show that the proposed method has the advantages of better performance,higher spatial resolution in azimuth estimation,and more accurate multi-target differentiation.The experimental data processing results verify the good engineering application adaptability and feasibility of the proposed method for engineering applications.3.The output post convolution joint processing and adaptive enhancement for the matched filtering in frequency domain using an acoustic vector sensor.The output signal model of the acoustic vector sensor in frequency domain after matched filtering is studied,which exhibits the same theoretical expression as that of the measurement model of the line spectrum signal in time domain.Therefore,it is proposed that the convolution joint processing in frequency domain is applied to the post-processing of the matched filtering output in frequency domain based on an acoustic vector sensor,in order to improve the detection performance of the known waveform wideband signal.The principle of the algorithm is given in theory.The simulation and experimental data processing results show that the proposed method has good detection performance and can improve the signal detection ability in engineering applications.4.Wideband signal azimuth estimation with known signal waveform using an acoustic vector sensor.According to the azimuth estimation principle of an acoustic vector sensor using histogram statistics based on complex sound intensity,wideband signals with known signal waveform can also be employed estimate the target azimuth by histogram statistics after matched filtering.The similarities and difference between the weighted histogram statistics azimuth estimation in frequency domain and the weighted histogram statistics in time domain are studied.Then,the advantages and disadvantages of uniform weighting,energy sum weighting,and the improved energy product weighted method are compared under different application conditions.Based on the output signal model of the matched filtering in frequency domain,a wideband azimuth estimation method based on the output using convolution joint processing post adaptive cancellation in frequency domain is proposed.The comparison of simulation and experimental data processing results show that the proposed method not only has the applicability in a lower SNR environment,but also can resist the impact of certain instantaneous strong interference,and effectively distinguish multiple targets.5.Experimental research on the application of active bistatic system based on acoustic vector sensor.Based on the theoretical model and the experimental model of a small active bistatic sonar detection system using an acoustic vector sensor as the receiver,the suppression of direct wave in the system,the echo output and echo detection under the condition of direct wave suppression are studied by experimental investigation.The distance estimation method,influence factors and the target distance estimation error in the system are derived.The effective performance of the system model,target detection and localization methods are verified by the comprehensive and systematic experiments,which provides an engineering reference for an acoustic vector sensor applied in the active bistatic sonar systems.This paper demonstrates the time reversal convolution joint processing using sound pressure and vibration velocity based on an acoustic vector sensor and its application in signal detection,the line spectrum signal azimuth estimation using post adaptive cancellation,the post convolution joint processing for the matched filtering output in frequency domain,the wideband signal azimuth estimation under the condition of known waveform,and the experimental investigation of the active bistatic detection system using an acoustic vector sensor.The research work provides a new technical scheme for the acoustic vector sensor application on small sonar platform.It can be widely used in marine monitoring,port security early warning,offshore defense and other fields,which exhibits good engineering research value and potential application prospects. |
PDF Full Text Request