Abstract: Wireless Sensor Networks (WSNs) are important in almost all applications which provide longer and low-cost supervising and motivating. In WSN applications, batteries are the main source of energy for sensor nodes where saving energy becomes more tedious. Today, almost every applications of WSN require sensor nodes which are unused to achieve fault tolerance and Quality of Service (QoS) during sensing. Similarly, the same unused nodes may not be necessary for multiple node communication because of the fewer loads in the traffic and the firm wireless links. This thesis presents a novel sleep-scheduling technique called Virtual Backbone Scheduling (VBS). VBS designed for WSNs consists of sensor nodes which are redundant. VBS forms multiple overlapped backbones which work alternatively to prolong the lifetime of the network. Here backbone sensor nodes only forwards the traffic, while rest of the sensor nodes turns off its radios to save energy. The alternative working of multiple backbones ensures that the entire network energy is distributed among all sensor nodes is balanced and thus a longer network lifetime can obtained against the existing techniques. Maximum Lifetime Backbone Scheduling (MLBS) is formulated as the scheduling problem of VBS. Approximation algorithms based on the Schedule Transition Graph (STG) and Virtual Scheduling Graph (VSG) is proposed. Theoretical analysis and simulation studies verify that VBS is superior to the existing techniques.
Keywords: Sensor nodes, WSN, VSB, STG, VSG.