Self-deployment of Mobile Agents in Manets for Military Applications

We present bio-inspired computation techniques, such as genetic algorithms, for real-time self-deployment of mobile agents to carry out tasks similar to military applications. Under the harsh and bandwidth limited conditions imposed by military applications, self-spreading of autonomous mobile nodes becomes much more challenging. In our approach, each mobile agent exchanges its genetic information, which is composed of speed and direction encoded in its chromosome (genome), with the neighboring nodes located in its communication range. A genetic algorithm run at the application layer as a software agent is used by each node to decide its next speed and direction among a large number of choices so that the unknown geographical area can be covered uniformly under conditions such as hostile attacks, natural (i.e., mountain, trees, lakes etc.) and man-made obstacles. We implemented a simulation software to quantify the effectiveness of the genetic algorithms under different military operational conditions (e.g., losing assets during an operation, the remaining agents should reposition themselves to compensate the lost in coverage and network connectivity). Metrics including normalized area coverage, deployment time, avoidance from obstacles over an unknown geographical area are used to demonstrate the efficiency of the self-deployment algorithm. The results show that genetic algorithms can be applied to autonomous mobile nodes and be performed as an effective tool for providing a robust solution for network area coverage under restrained communication conditions.