Research Of Knowledge-Aided Dim Target Trackingalgorithm

In modern warfare, with the application of low-altitude penetration technology，andthe development of maneuvering and stealth aircraft, the radar surveillance scenebecomes more and more complicated, meanwhile, the power of radar echo decreasessignificantly. Thus, tracking dim targets in complex environment has become achallenge for the modern radar system. Joint with the efficient information of targetsand environment, knowledge aided tracking, which improving the recognition of targetsand clutter, and reducing the interference of clutter, is able to provide a significanttracking performance with lower computation.Considering the above problems, in this thesis, the knowledge aided dim targettracking algorithm is studied. The specific contributions are introduced as follows:1. Focusing on the maneuvering target tracking, a modified IMM data associationalgorithm is given in this thesis. And it performs better transition rate of modelprobability，and also improves the tracking precision.2. In allusion to the utilization of amplitude information (AI) in complexenvironment, the mathematical formulations of the amplitude likelihood ratios ofdifferent clutter models have been deduced. By comparison with traditional algorithm,AI aided tracking algorithm has better tracking performance.3. Aiming at the interference of amount of false measurements, we study theDoppler information aided tracking algorithm. The algorithm has better data associationperformance.4. In order to tracking dim targets with complex terrain background, a trackmanagement algorithm in heterogeneous clutter has been proposed. Based on priorenvironment information, by adjusting strategy, this method has less false tracks andprovides better successful track probability.5. Directing at amount of false measurements, a track management algorithm basedon dynamic association is proposed. It produces better false track suppressionperformance.The algorithms mentioned above are tested by simulations. And the simulation results show the effectiveness of the knowledge aided dim target tracking algorithms incomplex environment.