Agents based control and coordination model for a distributed wireless sensor network to maximize field life.

Abstract

In recent years, wireless sensor network deployments for real life applications have rapidly increased. However, energy control remains one of the major barriers hindering complete exploitation of this technology because sensor nodes are usually powered by batteries with a limited lifetime. Therefore efficient energy management is key requirement for the success of a wireless sensor network. In this project, we proposed a distributed agent based model to solve the issue of energy management in WSNs. In our approach, sensor nodes are modeled as agents. One of the agents is a controller agent called the expert sensor agent (ESA). The ESA at the start of sensing sets a threshold energy level. During sensing, the ESA continuously monitors each of the other sensor agents’ energy levels and if a sensor node’s power level falls below the threshold, the ESA deactivates the node and at the same time activates one of the nodes currently in sleep mode. At the of the day, the sensing work load is equitably distributed across the nodes. The model was implemented using simulation experiments in OPNET. We evaluated our agent based model against none agent based technique in terms of power consumed and end to end delay (ETE). Our simulation results indicate that our agent based model achieves over 75% power savings compared to the non-agent techniques while at the same time ensuring real time delivery of data.