Location Tracking in a Wireless Sensor Network by Mobile Agents and Its Data Fusion Strategies

The wireless sensor network is an emerging technology that may greatly aid humans by providing ubiquitous sensing, computing and communication capabilities, through which people can more closely interact with the environment wherever they go. To be context-aware, one of the central issues in sensor networks is location tracking, whose goal is to monitor the roaming path of a moving object. While similar to the location-update problem in PCS networks, this problem is more challenging in two senses: (1) there is neither a central control mechanism nor a backbone network in such an environment and (2) the wireless communication bandwidth is very limited. In this paper, we propose a novel protocol based on the mobile agent paradigm. Once a new object is detected, a mobile agent will be initiated to track the roaming path of the object. The agent is mobile since it will choose the sensor closest to the object to stay. The agent may invite some nearby slave sensors to cooperatively position the object and inhibit other irrelevant (i.e. farther) sensors from tracking the object. As a result, the communication and sensing overheads are greatly reduced. Our prototyping of the location-tracking mobile agent based on IEEE 802.11b NICs and our experimental experiences are also reported.