Study On Array Signal Parameter Tracking Algorithm Based On Random Finite Set

The estimation of the Direction of Arrival(DOA)is the hotspot in array signal processing.The array signal will be interfered by noise during the propagation process,which will lead to inaccurate target estimation,and the DOA and number of signals will change with time.In recent years,DOA tracking technology under the framework of random finite set has made great progress.With the improvement of the development of the military surveillance systems,maneuvering target tracking technology has been becoming a hot.Interactive multiple model(IMM)method is an effective tracking of maneuvering targets on the research method,which can effectively improve the prediction accuracy of target state and avoid the failure of maneuvering target tracking caused by model mismatch.With the continuous expansion of target tracking theory,many maneuvering target tracking problems under non-standard measurement,such as interval measurement,array multi-signal superposition measurement,are worthy of attention.Based on the RFS theory,the main structure arrangement of DOA tracking method for array signals is as follows:1.This paper introduces the theory of random finite set and summarizes several filtering theories based on random set,which provides theoretical support for the follow-up work of this paper.2.For interval measurement of multiple maneuvering target tracking problem,the model concept is introduced of generalized labels multi-Bernoulli(GLMB)algorithm,and puts forward a kind of interactive interval measurement multi-model GLMB(IMM-GLMB)tracking algorithm in order to improve the efficiency of the algorithm,we proposed a fast algorithm adopted literature thoughts,successively put forward the interactive multi-model generalized labels Bernoulli fast(IMM-GLMBF)algorithm.IMM-GLMBF algorithm combines the prediction and update steps of IMM-GLMB algorithm,only needs a truncated.Compared with the IMM-GLMB algorithm,the proposed imm-glmbf algorithm is more accurate in estimating the number of targets and their states,which greatly reduces the computing cost.3.According to the problem of multiple source direction of arrival(DOA)tracking of acoustic vector sensor(AVS)array,a two-dimensional DOA tracking algorithm based on Multi-target Multi-Bernoulli(Me MBer)filtering is proposed.The MUSIC spectrum function has been regarded as the likelihood function of particle filtering and exponentially weighted to enhance the weight of particles at high likelihood area.The merits of thisproposal are that the Me MBer filtering is able to operate more directly on AVS array signals,effectively solving the problem of multiple time-varying target recognition.Simulation experiments show that this algorithm can accurately track the source state and estimate the number of sources.4.A two-dimensional DOA tracking algorithm based on multi-target multi-bernoulli filter is proposed for multi-source DOA tracking of acoustic vector sensor arrays.Because of the closed solution of the posterior probability density of multiple objects,the approximate solution is obtained by particle filter algorithm.The advantage of this method is that the member filter can process the AVS array signal more directly and effectively solve the problem of multi-time-varying target recognition.Simulation experiments show that this algorithm can accurately track the source state and estimate the number of source.5.According to the problem of multi-maneuvering targets Direction of arrival(DOA)tracking in acoustic vector sensor(AVS)array,we introduce the multi-model concept into the Multi-target Multi-Bernoulli(Me MBer)filtering algorithm and propose an Interactive Multi-Model Me MBer(IMM-Me MBer)tracking algorithm.The sampled particles are predicted by the Interactive Multi-Model(IMM)algorithm,then the predicted particles are updated by using the MUSIC spatial spectral function as likelihood function in conjunction with the Me MBer filter update strategy.With the combination of the characteristics of Me MBer filtering and the IMM method,this algorithm can effectively improve the target state prediction accuracy and avoid tracking failure caused by model mismatch in the maneuvering process of targets.Simulation experiments show that compared with Me MBer algorithm and PHD algorithm,the IMM-Me MBer algorithm can accurately track the target state and estimate the number of targets.