Barrier Coverage Mechanism Using Adaptive Sensing Range for Renewable WSNs

• Main Authors: Zaixiu Dong, Cuijuan Shang, Chih-Yung Chang, Diptendu Sinha Roy
• Format: Article
• Language: English
• Published: IEEE 2020-01-01
• Series: IEEE Access
• Subjects: Solar-powered sensor; probabilistic sensing model; adjustable sensing radius; barrier coverage; wireless sensor networks
• Online Access: https://ieeexplore.ieee.org/document/9087897/

Barrier coverage aims at constructing defense barriers to detect the intruder crossing the predefined boundary in a given wireless sensor network. In literature, most studies considered the battery-powered sensors and applied the Boolean Sensing Model (BSM). The battery-powered sensors have the constraint of limited lifetime while the BSM will affect the actual surveillance quality evaluation because it cannot reflect the physical features of sensing. This paper applies the Probabilistic Sensing Model (PSM) and proposes an algorithm, called BSAS, which considers the solar-powered sensors with adjustable sensing radius to construct the defense barriers. Two main challenges should be overcome. The first one is the cooperative working between sensors to achieve the highest intruder detection probability for a given boundary curve. The BSAS identifies the bottleneck segment with the minimal surveillance quality and schedules as many as possible sensors to improve the bottleneck segment. In addition, a space-time transformation scheme which further adjusts the sensing radius of some sensors is proposed, aiming at improving the detection probability of the bottleneck segment. Consequently, the minimal surveillance quality of the barrier can be maximized. The second challenge is to maintaining the perpetual lifetime of WSNs. The BSAS takes into account the recharging and discharging ratio and time length of daytime in a day in its energy management which guarantees that the sensor energy can always satisfy the energy consumption for sensor working even in the nighttime. Experimental studies reveal that the proposed algorithm outperforms the existing studies in terms of surveillance quality and stability.