Optimization of Mobile Sensor Movement in Self-Healing Sensor Networks

Abstract

This paper moves forward the key idea as proposed in past research works - a self-healing deployment approach for sensor networks, where a small percentage of mobile sensors are deployed along with the static sensors into a field of concern. Mobile sensors can move to make-up for a coverage holes or sensor failure and significantly boost network performance. However, since there are energy constraints on each individual mobile sensor, potentially receiving multiple requests from network holes, the decision to move a mobile sensor has to be optimum, one that maximizes network benefit. In this paper, I propose a hybrid distributed & central decision making algorithm to facilitate optimal moves by each mobile sensor. The algorithm uses several layered techniques like Rough Set analysis, sorting & multi-level auction to provide the best possible decision, given the network scenario and the approach is robust to incompleteness of information. The proposed solution also safeguards against network deadlocks and extensive simulations & statistical analysis have demonstrated superior performance of the algorithm when compared to its peers. Some traits of the algorithm proposed derive inspiration for decision support from Ants' swarm intelligence.