SNR-Adaptive Weighted Metropolis Consensus Filtering Algorithm for Distributed Target Tracking

This paper proposes a Metropolis consensus filtering algorithm with signal-to-noise ratio (SNR) dynamic weighting for distributed target tracking in sensor networks. Focusing on the challenges of tracking highly maneuvering aerial targets with nonlinear motion models and inefficient data fusion in non-uniform observation environments, we establish a novel SNR-based dynamic weighting model. The proposed framework incorporates real-time SNR evaluation of measurement data from each node and implements Metropolis-Hastings-based dynamic weight allocation for inter-node information exchange. Theoretical analysis proves the mean-square consensus convergence under time-varying network topologies. Simulation results demonstrate that during extreme 9g overload maneuvers, SNR-DWM-DCKF exhibits the most significant improvements: 8.41% higher accuracy and 7.14% faster convergence relative to DCKF. The proposed method establishes a theoretical framework and provides engineering implementation solutions for collaborative tracking of maneuvering target in complex electromagnetic environments.