A Framework for using Agents in Distributed Sensor Networks

Summary form only given. This paper describes a framework for using agents for multiple objective optimization in distributed sensor networks. A modular framework based on object oriented design is developed, which will facilitate the use of different agent based systems in simulations for detailed analysis of component interactions and the evaluation of the integrated system. This framework will be used to simulate different agent scenarios and generate data for performance analysis and comparison of these scenarios. The agent forms a wrapper around the node and perceives the environment from data collected by the sensors or messages from other agents or from events that occur within the node. Thus the agent has a measure of the state of the environment. Depending on the goal that has been set for the agent, the agent applies some condition-action rules. These rules help the agent to make a decision on what action has to be taken in order to achieve the goal. The agents in this system are used for target tracking, control and coordination of the sensor network, and reconfiguration of the system and its resources in order to maximize the field life. These intelligent agents will also be used to increase the ability of the field to survive attacks, failures, and accidents. Thus using this framework, different agent based scenarios can be easily implemented. Some of the scenarios that can be explored are the use of single, multiple, static, mobile, autonomous, and social agents. The data generated from the simulations will be used for comparison analysis of the different agent scenarios which will help in designing an improved agent based distributed sensor network system