Abstract:
Wireless sensor networks have many applications. In our research, we are particularly interested in applications of monitoring a field to detect emergent incidents, such as forest fire, or intrusion objects such as enemy tanks, vehicles, submarines, etc. The traditional way for environment monitoring is to deploy sensors all over the region. However, this way not only wastes huge amounts of money for sensor nodes and resources, but it is also unnecessary and unrealistic sometimes. It provides huge amounts of garbage in the form of dead sensor nodes and batteries. It invokes many data collisions and places a serious burden on network protocols. Since sensor nodes have limited energy supply, sensor networks may be configured to put some sensor nodes in sleep mode to save energy. We first studied the deployment and schedule strategy of sensors. We also study the impact of object size for intrusion detection. In many applications, we have no need to have detail on every point inside the area, but only in some critical places. For example, in a forest fire surveillance application, covering the whole forest is unnecessary and unrealistic. Then, we propose a divide and conquer based environment surveillance methodology. We deploy sensors along critical lines instead of all over the region which is divided into some small area. Our aim is to detect, or even capture a target or target event rather than tracking it in real time. There are main three components in the design: sensors, sinks, and robots. Instead of deploying sensors all over the monitored area we only deploy sensor in some critical areas which divide the interested region into small areas. In this strategy, robots play critical role for handling events. Next, we propose bio-inspired strategies for large area surveillance based on wireless sensor networks. Inspired by the scent-marking behavior among primates and many other mammals, we propose and study a collaboration strategy for mobile and static sensors. Finally, we propose a digital pheromone based patrolling strategy of robots.
