Weak Signal Detection And Tracking Based On Improved Particle Filter

In recent years, detection and tracking technology of the weak signal has been employed in various fields, such as industry,transportation and national defense. But along with the improving of precision, challenges of signal to noise separation are also increasing highlighted. Based on this, detection and racking method based on improved particle filter is proposed and the accurate tracking for the weak signal is implemented which in low signal-to-noise ratio.First, the advantages and disadvantages of track before detect (TBD)and detect before track (DBT) was explained, the observation model of passive sensors and the model of uniform linear motion are established and a detailed introduction to the Bayesian estimation and particle filter theory was given in this paper. The superiority of particle filter processing before detection is given for the follow-up study.Secondly, the basic particle filter track before detect algorithm is introduced, and the pre-detection algorithm based on the traditional particle filter is used to verify the uniform motion model. Based on the traditional improved algorithm, the Markov chain quasi-Monte Carlo algorithm is used to modify the resampling process. Then, an improved Quasi-Monte-Carlo intelligent particle filter track before detect is proposed. The algorithm improves to a certain extent. The problem of lack of particle diversity and to a certain extent, reduce the running time of the algorithm. The simulation results show that the improved Monte Carlo intelligent particle filter algorithm has improved the efficiency and tracking accuracy.Finally, based on the improved Monte Carlo intelligent particle filter algorithm, the modeling method of Interacting multi model is proposed for the detection and tracking of dim targets and an improved interactive multiple model algorithm is proposed to optimize the model.The simulation results show that the interactive multi model algorithm can reduce the number of particles to a certain extent under the premise of ensuring the tracking accuracy, and accurately tracking the weak targets of uniform acceleration and turning motion. It proves the validity and reliability of the improved Quasi Monte Carlo Intelligent Particle Filter Algorithm for dim target detection and tracking.