High-precision DOA Estimation Of Underwater Targets Based On Ultra-low Frequency Vector Processing

The acoustic vector sensor is composed of an omnidirectional pressure sensor and a particle sensor.It can simultaneously measure and pick up the information of pressure and particle velocity in acoustic field.Compared to a conventional pressure sensor,an acoustic vector sensor can detect acoustic vector information including information of spatial orientation.Since acoustic vector sensors were invented,there are a large number of direction-of-arrival(DOA)estimation methods based on vector hydrophones.Thereinto,the pressure-velocity combined processing technology on the basis of concept of acoustic energy flux has a good anti-noise ability about isotropy,which is attracted extensive attention and researches by scholars from home and abroad.However,in actual marine environment,there are a large number of asymmetrically distributed noise sources.These asymmetrically distributed noise sources make the ocean environmental noise field appear to be anisotropic.Because anisotropic noise cannot be suppressed by conventional measures,it seriously interferes with the DOA estimation accuracy based on vector hydrophone.For these problems above,the paper reviews the vector acoustic field theory,physical basis of the pressure-velocity combined processing technology and the DOA estimation method based on this technology,which combines anisotropic noise suppression method with pressure-velocity combined processing technology.Analyzing anisotropic noise field model as a starting point,discussion about anisotropic noise field to interference principle of the DOA estimation accuracy based on the vector hydrophone is made,which further makes full use of vector information detecting from vector hydrophone to propose an acoustic vector compensation method in order to achieve suppression of anisotropic noise.This provides a new idea and method for solving the problem of high precision DOA estimation in anisotropic noise field under low SNR conditions.The specific steps of this thesis are as follows:(1)In order to solve the problem that anisotropic noise interferes with the accuracy of DOA estimation,a method for suppressing anisotropic noise at ultra-low frequency is proposed.Combined electronic rotation with acoustic vector compensation method,this reduces the interference of nontarget noise source's radiation noise against target radiation noise so as to improve the purpose of the DOA estimation accuracy.(2)The paper compares the computer-aided simulation of the high-precision DOA estimation method of underwater targets based on vector processing,conventional complex intensity processor and ESPRIT algorithm on the basic of a single vector sensor,which verifies the algorithm performance.(3)Based on the DSP chip design,the high-precision DOA estimation system of the underwater target was achieved and the performance of DSP system was verified by simulating real input condition.The result of simulation showed that the system can perform DOA estimation on multiple underwater targets in an anisotropic noise field with high precision.