Optimization of Watchdog Selection in Wireless Sensor Networks

Abstract

Application areas of wireless sensor networks (WSNs) are rapidly expanding these days. This also intensifies security concerns in extensively deployed WSNs. A watchdog system is one of the security enhancement methods. In such a system, a number of sensor nodes are selected as watchdogs that monitor their single hop neighbors. Thus, sensing operations lose resources to combat distrust. This letter develops models that optimize the selection of watchdogs in WSNs. It focuses on two major facts: 1) overlapping and 2) coverage. Overlapping occurs when a sensor node is monitored by multiple watchdogs. It causes additional consumption of resources. It is inevitable due to the propagation characteristics of wireless signals. The full coverage occurs when each sensor node in a WSN is either monitored by at least one watchdog or working as a watchdog. This letter presents three optimization models for watchdog selection in WSNs. It also evaluates the models through case studies for realistic WSN topologies. The presented models provide a better understanding of resource efficient watchdog deployment strategies.