Research on target attributes fusion in netted-radar based on d-s evidence theory
Yu Juan Yu Juan, Zhang Bo-yan Zhang Bo-yan, Yu Ji-zhou Yu Ji-zhou · 2015
The information provided by netted-radar usually contains much more uncertainty, and counterintuitive results would be generated by D-S evidence theory with highly conflicting evidences. A model-modified D-S method is proposed in this paper, which can acquire rational fusion results when the evidences are highly conflicting. In addition, a two-level fusion structure is put forward to further eliminate the uncertainty of single-scan measurements. Simulation results show that the novel method could effectively weaken the conflict between evidence, and improve the reliability of fusion result which reduces the decision-making risk.