Energy Efficient Clustering in Sensor Networks with Mobile Agents

Wireless sensor networks with mobile access points are effective tools to collect data in a variety of environments. Low-cost and low-power sensors in the reachback operation contend for the channel to transmit their own data packets to the mobile agent. This data communication should be designed to ensure energy efficiency and low latency. In this paper, we propose a clustering scheme for wireless sensor networks with reachback mobile agents (C-SENMA) toward that goal. C-SENMA groups sensors into clusters such that nodes communicate only with the nearest clusterhead (CH) and the CH takes the task of data aggregation and communication with the mobile agent. In our scheme, CHs use a low-overhead medium access control (MAC) mechanism very similar to the conventional ALOHA to contend for the channel. Using results from random geometry theory, we analyze the clustering performance under the realistic MAC algorithm. Our analysis enables us to obtain the optimal average cluster size which minimizes energy consumption. We justify our analysis results by extensive simulations according to various clustering parameters. Furthermore, we study the effect of underlying physical layer characteristics on the amount of energy reduction achievable by the proposed clustering architecture