Abstract

Wireless sensor networks consist of a large number of small battery powered sensor nodes with limited energy resources which are responsible for sensing, processing, and transmitting the monitored data. Once deployed, the sensor nodes are normally inaccessible to the user, and thus replacement of the battery is generally not feasible. A major concern in designing and operating dense Wireless Sensor Networks (WSNs) is the energy-efficiency. Hierarchical clustering and cross-layer optimization are widely accepted as effective techniques to ameliorate this concern. We propose two different novel energy efficient algorithms to gather data from sensor nodes. Energy-Efficient Media Access Control (EE-MAC) protocol is the first algorithm, which has excellent scalability and performs well for both small and large sensor networks. We will also provide a theoretical analysis of the protocol and give guidelines on how to find the optimal protocol parameters such as the number of clusters. In addition, we develop and analyze a novel and scalable Spiraled Algorithm for Data-gathering (SAD) that periodically selects cluster heads according to their geographic locations and residual energy by sorting nodes on virtual spirals. Theoretical analysis and simulation results show that SAD can achieve as much as a factor of three prolonging network lifetime compared with other conventional protocols like LEACH especially when the network is large. Moreover, SAD is also able to distribute energy dissipation evenly throughout the sensors such that 80% of the nodes run out of batteries in the last 20% of the network lifetime.