Multiple Unknown Emitters Direct Tracking With Distributed Sensor Arrays: Non-Homogeneous Data Fusion and Fast Position Update

This paper considers the problem of multiple unknown emitters tracking based on distributed sensor arrays. The most common two-step methods firstly track the intermediate parameters and then solve emitter positions. However, the two-step methods are suboptimal as ignoring the matching information. In this paper, we introduce the novel thought of Direct Position Determination (DPD) to the tracking problem and propose the Non-homogeneous Data Fusion and Fast Position Update (NDFFPU) algorithm. Firstly, the NDFFPU algorithm adapts to tracking scenarios by reducing the weights of past data in autocorrelation estimation. Thereafter, the non-homogeneity error of received data is eliminated in the difference co-array domain, which reduces the sensitivity to observation position. Finally, the NDFFPU algorithm fuses all difference array data and performs the Newton iteration to update the emitter positions with high speed. The NDFFPU algorithm avoids the drawbacks of both the intermediate parameter estimation and the complex peak-searching. Extensive numerical simulations demonstrate the superiority of the NDFFPU algorithm in utilized array aperture, computational complexity, tracking accuracy, and robustness.

Automated summary

This paper proposes an algorithm for tracking multiple unknown emitters based on distributed sensor arrays. The algorithm introduces the novel thought of Direct Position Determination (DPD) and improves tracking performance through Non-homogeneous Data Fusion and Fast Position Update (NDFFPU). Numerical simulations demonstrate the superiority of the algorithm.