Study Of DOA Estimation Of Multiple Sources Based On The Towed Fiber Optic Vector Hydrophone Array

Due to the high sensitivity and the ability to solve the port/starboard discrimination problem,etc.,the key technology of the fiber optic vector hydrophone towed line array has become a research hotspot.Because a large number of targets exist in the water and the surface of the water,DOA estimation of multiple sources need to be solved in array application.The classical spatial spectrum estimation algorithms require large snapshots and high signal to noise ratio,and perform poor when sound sources are relevant.Compressed sensing theory utilizes space sparsity of sound sources,to realize high resolution DOA Estimation and provide a new way for the vector array multi-object orientation.The paper focus on multi-target orientation based on the vector array,the technology of fiber optic vector hydrophone towed array and high resolution direction algorithms based on vector array were studied.The main contents are as follows:(1)The design of towed fiber optic vector hydrophone array was studied.The theory of miniaturized fiber optic hydrophone used for towed array was studied,and basic properties of the developed fiber optic hydrophone were measured.Effect on target orientation was studied because of the center inconsistent between velocity sensor and pressure sensor when fiber optic hydrophone is assembled in practical.In working frequency band 80~2000Hz,simulation results show that if the two sensor spacing less than 0.2m,think is concurrent in water.Attitude control of optic vector hydrophone was studied through simulation,the results show that if the roll angle and pitch angle change less than 10 degrees,the impact on the goal orientation is small.Moreover,the attitude control method for towed hydrophone arrays was given.(2)CBF,MVDR,MUSIC methods were studied on multi-source orientation of the vector array.The signal model of single vector sensor and vector array signal model are given,and the performance of these three methods in vector array signal model is simulated and analyzed.The results showed that three directional algorithms can solve the port/starboard discrimination problem.CBF has low resolution,MVDR and MUSIC get higher resolution under condition of incoherent signal sources and high signal-to-noise ratio.However when signal sources are relevant or the number of snapshots is small,the directional performance will decline.(3)The basic theory of compressive sensing is studied,and the sparse signal model of vector array and pressure array is constructed respectively.Two methods of reconstructing sparse signal are studied.Directional performance of the these two algorithms under the pressure hydrophone array and vector array sparse signal model was analyzed through simulation.The results show that compressed sensing algorithm can resolve multiple targets with single snapshot,but when the signal-to-noise ratio is low,false peaks and results orientation bias appear.Moreover,the performance of the proposed algorithm is not decreased in the condition of multiple coherent sources and small snaps.