Random field estimation with quantized measurements in sensor networks

摘要

In this paper, we study the sensor selection problem for random field estimation in sensor network. The combinatorial optimization problem of choosing sensor nodes to minimize the estimation error with bounded observation cost is proved NP-complete. Iterative algorithms is proposed to pursue the suboptimal solution for the sensor selection problem under the D-optimal criteria. Furthermore, the optimal estimator for the Gaussian random field with quantized measurements is proposed in the minimum error variance sense. Simulation results show the performance of our proposed algorithms.