E-DSR: energy-efficient routing for sensors with diverse sensing rates

Cluster-based routing is popularly used in wireless sensor networks (WSNs), where sensors are organised into clusters and cluster heads (CHs) are selected to compress and forward packets for other nodes. However, most of existing protocols implicitly assume that sensors produce data with the same speed. Due to event occurrence or application needs, sensors may have different sensing rates in practice. Some CHs may thus encounter serious buffer overflow and dispose of many packets. To conquer this problem, the paper proposes a protocol called energy-efficient routing for sensors with diverse sensing rates (E-DSR) to extend network lifetime and diminish lost packets. E-DSR divides the network into grids and selects one CH in each grid based on multiple factors such as its position, residual energy, and sensing rate, so as to improve energy efficiency on routing. Moreover, depending on traffic loads of CHs, E-DSR adaptively splits or merges grids to avoid buffer overflow or facilitate data compression, respectively. Simulation results verify that E-DSR significantly prolongs network lifetime and reduces the data loss rate, as compared with various routing protocols developed for WSNs.