A fault-tolerant sensor scheduling approach for target tracking in wireless sensor networks

Abstract

Target tracking is a critical application in wireless sensor networks (WSNs). Existing target tracking algorithms basically perform node scheduling based on trajectory prediction. Once the target is lost due to prediction errors, the target recovery mechanism performs a search to find the target, which may require the system to activate a large number of nodes and cause additional energy consumption. Moreover, the target data may be lost due to the time consuming of target recovery. To solve this problem, we propose a fault-tolerant sensor scheduling (FTSS) approach to reduce the target loss probability as much as possible. Furthermore, we design a low-power scheduling mechanism in FTSS to reduce energy consumption. In FTSS, we first design a fault-tolerant domain to expand the scheduling range of candidate nodes. Then, we consider remaining energy, sensing coverage, and overlapping coverage to activate as few sensors as possible to cover the fault-tolerant domain. Moreover, we employ an improved binary Grey Wolf Optimizer (bGWO) in the optimization process to speed up the convergence. Numeric evaluations indicate that, compared to DPT, HCTT, GBRHA, and E2DR-MCS, our method achieves 11%, 2.3%, 7.8%, 4.6% reduction in target loss probability, and improves network lifetime by 138%, 49%, 31%, 19%, respectively.

Keywords

• Wireless sensor networks
• Target tracking
• Sensor scheduling
• Prediction errors
• Target loss
• Binary grey wolf optimizer