An Improved Multi-model Particle Filter Track-Before-Detect Algorithm

Abstract

In the field of weak target detection and tracking, when the target maneuvers, the detection capability of the traditional Multiple-Model (MM) Particle Filter (PF) Track-Before-Detect (TBD) algorithm decreases and the calculation amount of the algorithm increases. In this paper, an improved MM-PF-TBD algorithm is proposed. By adding the Constant Acceleration (CA) model, the Coordinate Turn (CT) model with variable turning rate and the Quasi-Monte Carlo (QMC) method, the improved algorithm could achieve the detection and tracking of weak maneuvering targets. Simulation results show that the detection capacity of the improved algorithm is 15% higher than that of the traditional algorithm, and the improved algorithm has more stable performance in low Signal-to-Noise Ratio (SNR) environment.