An Energy-Balanced and Timely Self-Relocation Algorithm for Grid-based Mobile WSNs

Abstract

The sensor relocation problem (SR problem) is an important issue in wireless sensor networks (WSNs). The main difference between SR problem and sensor deployment (SD) problem is that SR problem has a strict response time requirement. Because the SD problem occurs during network initialization, so there is no strict response time requirement. The SD problem is solved as long as the required coverage level is met. However, the SR problem occurs during operation of WSNs, we need to minimize the impact on the surveillance task while replacing faulty sensors with redundant sensors. In this paper, we revisit the SR problem in mobile WSNs. In terms of response time and moving distance, the proposed algorithm moves redundant sensors to faulty sensors and coverage holes by cascaded movement. It also considers the estimated value of network lifetime in the planning of cascading schedule. Hence, the SR problems caused by inappropriate cascading schedules can be avoided. Overall, the proposed SR algorithm has the following features: 1) shorter response time; 2) shorter moving distances; 3) smaller number of requesting places; and 4) longer network lifetime. The performance evaluation done in this paper has confirmed its superior performances in all these aspects.