Multi-sensor Detection and Estimation Fusion by Minimizing Improved Minkowski Distance
Wenyan Zhu, Yong Chang, Peng Chen, Qingfang Teng · 2023
In this paper, we focus on the problem of multi-sensor joint detection and parameter estimation fusion under inaccessible prior probability, in which numbers of sensor nodes collaborate to make a decision between two hypotheses and estimate the unknown parameter and the guess of true prior associated with the decided hypothesis simultaneously in the fusion center. Firstly, we derive the multi-sensor joint likelihood function under either hypothesis by the property of the products and convolutions of Gaussian probability density functions. Then, the optimal decision rule is obtained by building and solving an unconstrained optimization model that maximizes the weighted combination of the expected utility and the average detection probability. Finally, an improved Minkowski distance measure method is given to choose the optimal guess of the true prior probability. Simulation results show the superiority of the proposed approach.