Optimal bit allocation for maneuvering target tracking in UWSNs with additive and multiplicative noise

摘要

Due to the limited energy and bandwidth in underwater wireless sensor networks(UWSNs), the original measurements should be quantized before transmitted to the fusion center. In this paper, the optimal bit allocation is utilized to improve the tracking accuracy on the premise of transmitting the same number of quantized bits of data. The relationship between the quantization level of sensor nodes and posterior Cramer-Rao lower bounds (PCRLB) is derived and taken as the performance bound for tracking accuracy. Then the problem of optimal bit allocation is converted into an optimization problem. To make computation-efficiency, allocating the maximum bits for each candidate sensors and then the generalized Breiman, Friedman, Olshen, and Stone (GBFOS) algorithm is adopted to delete one bit at a time until the total number of bits is satisfied. In addition, the genetic algorithm for bit allocation (GABA) is proposed in this paper to solve the optimization problem when the transmission bits and the number of candidate sensors are large.The simulation results illustrate the performance of the proposed scheme in improving the tracking accuracy on the condition of limited communication bandwidth. Compared to GBFOS, GABA proposed in this paper can satisfy the real-time target tracking requirements and ensure tracking performance. (C) 2019 Elsevier B.V. All rights reserved.