Energy-aware distribution of monitoring agents using Genetic Algorithms

In a mobile agent-based monitoring network, mobile monitoring agents travel over the network to detect abnormal conditions in real-time. Each mobile monitoring agent is a damage diagnosis program which can recognize certain type of damage. To achieve effective damage detection and prolong the lifetime of a monitoring network, this paper studies energy-aware distribution of mobile monitoring agents using Genetic Algorithm (GA). The goal of the GA-based algorithm is to distribute monitoring agents to sensor nodes where sensor feature vectors have high affinities with memory cells of a monitoring agent and the remaining battery capacity of the sensor node is above a predefined threshold. With this distribution strategy, the system is able to send certain type of monitoring agents to the site where they are needed. When there are more than one sensor nodes in the suspicious region, the sensor node with high remaining battery capacity will have higher priority to receive a mobile monitoring agent. This distribution scheme makes the remaining battery capacity of sensor nodes in the network more balanced so that the lifetime of the monitoring network is extended.