Distributed adaptive scheme for reliable data collection in fault tolerant WSNs

Abstract

Reliable data packet transfer is an essential facet of quality of service in Wireless Sensor Networks (WSNs). WSNs require pertinent and reliable data collection schemes to provide dependable information. We analyze the factors that affect reliability, and design efficient combinations of the possible options. While many researchers have tackled this issue by focusing on particular layers of the embedded communication stack (e.g., Medium Access Control (MAC) or routing layer), we propose a layer-independent transmission rate adaptive scheme, where no additional control messages are required. We focus on packet loss avoidance along prone-to-failure paths to the sink with the main objective being the efficient queue management of a sensor node. Based on certain factors (e.g., application or queue requirements) nodes will consequently take adjustment decisions. Our thorough simulation experiments demonstrate that DRAS reduces the packet drops and improves the data collection procedure in an energy efficient manner.