Using Ship Radiated Noise Spectrum Feature for Data Association in Underwater Target Tracking

In multiple-target tracking problem, data association technique plays an significant role. When targets move closely or crosswise, performances of conventional data association algorithms which use kinematic information only may be degraded. Actually, beside the kinematic information, sensors always can obtain feature information about the target, and incorporating the features into data association problem can improve the performance of data association. In this paper, for underwater target tracking, we propose a ship radiated noise spectrum feature aided probabilistic data association (PDA) algorithm to improve the data association and target tracking performance. Simulations are finally carried out and results show that the proposed method can improve the tracking performance over the conventional PDA algorithm in close-space-targets scenario.