Hole Detection and Healing in Hybrid Sensor Networks

TL;DR

This paper presents centralized algorithms for detecting and healing coverage holes in hybrid wireless sensor networks, utilizing Voronoi diagrams and greedy algorithms, with simulations demonstrating effective hole coverage by mobile sensors.

Contribution

It introduces novel centralized algorithms for precise hole detection and area maximization in hybrid WSNs with obstacles, including a polynomial-time boundary identification method and a greedy approximation approach.

Findings

Voronoi diagram-based boundary detection accurately identifies holes.

The greedy algorithm achieves a 50% approximation ratio for coverage maximization.

Simulations confirm effective hole coverage by mobile sensors.

Abstract

Although monitoring and covering are fundamental goals of a wireless sensor network (WSN), the accidental death of sensors or the running out of their energy would result in holes in the WSN. Such holes have the potential to disrupt the primary functions of WSNs. This paper investigates the hole detection and healing problems in hybrid WSNs with non-identical sensor sensing ranges. In particular, we aim to propose centralized algorithms for detecting holes in a given region and maximizing the area covered by a WSN in the presence of environmental obstacles. To precisely identify the boundary of the holes, we use an additively weighted Voronoi diagram and a polynomial-time algorithm.Furthermore, since this problem is known to be computationally difficult, we propose a centralized greedy 1/2-approximation algorithm to maximize the area covered by sensors. Finally, we implement the algorithms and run simulations to show that our approximation algorithm efficiently covers the holes by moving the mobile sensors.