RCS Information-Aided Distributed Multisensor LMB Filtering Algorithm

In complex electromagnetic environments, existing multi-target tracking algorithms often suffer from performance degradation. To address this challenge, this paper proposes a Distributed Multi-Sensor LMB algorithm based on RCS information assistance. The algorithm models target RCS information parameters using a non-stationary autoregressive Gamma process to achieve Bayesian RCS estimation. This method comprises two main steps: single-node local filtering and multi-node information fusion. During the local filtering phase, the target state is first extended by introducing RCS information parameters to establish the likelihood function, thereby deriving updated prediction and update equations. This process fully utilizes the target's RCS information to enhance the distinction between targets and clutter. Subsequently, based on this information, we establish a label matching quality model and combine label historical data to achieve label matching among multiple nodes. This matching process enables effective fusion of multi-sensor data, thereby improving overall tracking accuracy. Simulation experiment results demonstrate that the proposed method exhibits strong effectiveness and robustness in dense clutter environments and under low detection probability conditions.