Research On Techniques Of Data Association In Complicated Environment

Radar target detection and tracking has been an important research topic. In recent years, radar signal processing under the sea clutter attracts the attention of a growing number of scholars.This thesis studies the target tracking under sea clutter. In this thesis, sea clutter signal as background noise, and focus on the introduction of the measurement target amplitude characteristic information, which improving the accuracy of the target state estimation and tracking performance.With the increasing development of science and technology matures, gradually increasing of the resolution of radar, sea clutter performances non-stationary characteristics under the radar measurement. Accurately estimating of the target under the background clutter characteristics is non-smooth has become an important research topic. If we can make full use of radar received measurement information, then it’s more possible to obtain accurate target state. So, this thesis will echo amplitude information into target tracking algorithm to improve the estimation accuracy of maneuvering target state. Under the irradiation of the high-resolution radar, Rayleigh distribution model can’t be used to exact fit the sea clutter, because of the non-stationary of sea clutter, which seriously affect the tracking performance of the maneuvering target. In order to obtain a better target state estimation at a relatively low computational complexity, this thesis uses K distribution model for sea clutter modeling, and combines with probabilistic data association algorithm and particle probability hypothesis density filter algorithm under the single-target tracking and multi-target tracking studies.In this thesis the main work has the following several parts:The first part, the thesis studies the formation mechanism and amplitude characteristic of sea clutter, definite the spatial and temporal correlation of sea clutter. Given the size of several common statistical model and the model of the distribution of probability density function of the graphics.The second part, the thesis studies the maneuvering target tracking technology under the sea clutter. Firstly, we summarize the traditional research of maneuvering single target, and then using probabilistic data association(PDA) algorithm for target state estimation. Based on PDA algorithm, we focus on the introduction of the improves algorithm based on measurement amplitude information, and analysis the simulation of the algorithm.The third part, the thesis studies the sea clutter of maneuvering multi-target tracking algorithm. Due to multiple target tracking, data association algorithm of target and track data need to be associated, large amount of calculation and difficult to close movement target tracking effectively. Therefore, the thesis introduces the probability hypothesis density(PHD) algorithm, the distribution of K filtering estimation of target in clutter. Several algorithm in the estimation of the target is relatively optimal performance.