Research On Multipath Echo Processing Method Of Wideband Moving Target For Active Sonar

Active sonar is currently the main method of underwater target detection and environmental measurement,and it is widely used in marine development and national defense.Especially with the continuous development of noise reduction technology,active sonar becomes more important in long-distance underwater target detection.Wideband signals can obtain higher matching processing gain and range resolution based on the principle of pulse compression.However,many factors cause the signal distortion,thus affect the matching processing performance seriously,such as ocean multipath channel and time-frequency spreading of moving-scale target.The Doppler sensitivity of wideband signals is weak,which reduces the velocity estimation performance.Therefore,the improvement of processing performance for wideband moving-scale target echo has always been a research hotspot and difficulty in underwater acoustic field.In this paper,we propose a variety of processing methods for wideband distortion target echoes to improve the detection and estimation performance under multipath channels from various aspects of research,such as optimization of transmitted signal,echo processing of moving-scale target,matching reception under spreading channel,and estimation of channel impulse response.The main research contents are as follows:(1)This paper analyzes the resolution performance of active sonar emission signals,and provides a theoretical basis for the optimal waveform parameter selection and structural design.Based on the equivalent relationship between the ambiguity function and the matching correlation,the performance of the sonar signal when using different detection methods is analyzed through the narrowband,wideband and long-time ambiguity functions,and the matching results of the distorted echo in practical applications can be predicted.The research results show that the wideband detection performance of wideband Doppler signals is better,for example,HFM signal with Doppler invariance is very suitable for moving target detection while there is a time-frequency coupling phenomenon,and the influence of acceleration on long pulse width signal cannot be ignored during detection.Accordingly,multiple combined signals are often used in active sonar,thus this paper applies the ambiguity function calculation that takes frequency,bandwidth,pulse width,number of repetitive pulses,and intra-pulse modulation as inputs,considering the time and frequency resolution,to clarify the quantitative evaluation method for active sonar signal optimization.(2)This paper analyzes the spreading scattering characteristics of moving-scale targets,and investigates the time-frequency diversity detection methods,then proposes a velocity measurement method for wideband multi-highlight targets.The plate element method is used to physically describe the scale target,and an equivalent wideband echo model of multihighlight target is given.According to the characteristics of the echo signal,two matching detection methods based on waveform diversity and extended characteristic diversity are applied to achieve effective detection of long-time widewidth echoes and to combat the timefrequency spreading caused by target scale and velocity.The velocity measurement performance of wideband multi-highlight echoes becomes worse because of the aliasing of multi-reflections,thus a novel target velocity measurement method for HFM multi-highlight echo is proposed.This method dynamically compresses and loads a variable-length time window to the echo,to increase the energy proportion of the main reflection,thereby improving the velocity measurement performance of aliasing echo.(3)This paper investigates the echo detection methods in multipath Doppler channel,and proposes an estimation method of channel impulse response that can be applied to low frequency wideband LFM signals in low SNR environment.Based on the ray acoustics theory,a sparse model of the multipath Doppler channel is established.In view of the distortion of wideband signals in multipath Doppler channels and the difficulty of the estimation of channel impulse response under low SNR conditions,this paper proposes an iterative Fr FT-based method to estimate the channel parameters,and the sensitivity of Fr FT to the LFM signal modulation rate is used to estimate the Doppler factor and the corresponding time delay,and then the copy signal in the matching processing is predistorted by the estimation result of channel impulse response.Finally,the echo detection performance under spreading channel is improved even if the channel spreading is unknown.Finally,based on the high-speed digital signal processor platform,we design and implement a real-time active sonar experimental system for target detection and parameter estimation.The system is divided into pre-processing,echo detection and target parameter estimation.In the echo detection process,when the target spreading information is known,the appropriate transmitted signal can be selected according to the signal resolution unit and the echo signal can be processed in a diversity method.When the target spreading information is uncertain,the echo signal can be processed by diversity within the range of target velocity and scale extension.When the channel spreading information is known,the optimum matching method of RC,RCI,SRC and TFRC can be selected correspondingly.When the channel spreading information is uncertain,the receiving method of echo matching between frames can be selected,or the echo correction matching can be performed by the estimation of channel impulse response.The target characteristic estimation includes a series of target parameters such as radial velocity and scale,where the velocity estimation generally uses the wideband matching measurement method.The transmitted signal with special signal form can use the corresponding velocity measurement method.Finally,an example is used to verify the real-time and correctness of each functional modules of the experimental system.