Pheromone-based In-Network Processing for wireless sensor network monitoring systems

Abstract

Monitoring spatio-temporal continuous fields using wireless sensor networks (WSNs) has emerged as a novel solution. An efficient data-driven routing mechanism for sensor querying and information gathering in large-scale WSNs is a challenging problem. In particular, we consider the case of how to query the sensor network information with the minimum energy cost in scenarios where a small subset of sensor nodes has relevant readings. In order to deal with this problem, we propose a Pheromone-based In-Network Processing (PhINP) mechanism. The proposal takes advantages of both a pheromone-based iterative strategy to direct queries towards nodes with relevant information and query- and response-based in-network filtering to reduce the number of active nodes. Additionally, we apply reinforcement learning to improve the performance. The main contribution of this work is the proposal of a simple and efficient mechanism for information discovery and gathering. It can reduce the messages exchanged in the network, by allowing some error, in order to maximize the network lifetime. We demonstrate by extensive simulations that using PhINP mechanism the query dissemination cost can be reduced by approximately 60% over flooding, with an error below 1%, applying the same in-network filtering strategy.