A Mobile Sensors Dispatch Scheme Based on Improved SOM Algorithm for Coverage Hole Healing

Abstract

The coverage holes caused by sensor failures severely degraded quality of service (QoS) in the sensing layer of Internet of Things (IoT). To improve the reliability and QoS of sensing layer, coverage hole healing problem has drawn considerable attention among researchers. Coverage holes healing problem aims to select and dispatch a set of mobile sensors to coverage holes detected by static sensors. In this paper, an improved Self-Organized Mapping (SOM) is proposed for coverage holes healing in hybrid sensor network composed of static sensors and mobile sensors. Firstly, the fuzzy system is introduced to optimize the selection strategy in SOM. The improved strategy assigns eligibility to each mobile sensor, which is calculated by the moving distance of mobile sensors and the residual energy of mobile sensors. The mobile sensor with the highest eligibility is selected. Secondly, a priority decision model is designed to measure the importance of failed sensors in sensor network. This model analyzes the location information and energy factor of failed sensors. Finally, according to failed sensors with different priorities, a new learning parameter is designed. The proposed scheme can adaptively update the weights of the winner in SOM and it can efficiently dispatch mobile sensors while maintaining the reliability of the network. Simulation results demonstrate that the proposed scheme can obtain better performance than the existing algorithm in aspects of efficient coverage hole healing, energy consumption, and network lifetime.