Comparative Analysis of Time Synchronization Methods for Wireless Sensor Networks

Abstract

Wireless sensor network usually consists of very large number of nodes having limited resources in terms of memory, power and processing capabilities. It is therefore necessary to use the available resources in an efficient manner to prolong the lifetime of the network. A typical wireless sensor network faces many challenges during its design, due to scarcity of energy supply in sensor networks. So, unlike other wireless networks protocols which are aimed to maximize capacity, lower bit error rate etc. the protocol design here in wireless sensor networks need to focus on efficient use of available energy within the network. One of the major concerns in wireless sensor networks design is time synchronization. As the sensor nodes have limited energy resources and once deployed may remain unattended for a long period of time, so for the entire network to work in coordination it is important to maintain synchronization between all the nodes in the network. In this dissertation we compare two existing time synchronization schemes namely Reference Broadcast Synchronization and Adaptive Clock Synchronization for different network environments and present an algorithm which provides network wide synchronization while assuring the minimal energy dissipation. In a sensor network every time a synchronizing node in a region dies, all the receiver nodes under its vicinity are dropped which affects the network coverage and connectivity of that region in an adverse manner. Thus re-synchronization of the entire network is necessary, which is an inefficient procedure for large networks as it consumes a lot of energy. The proposed algorithm instead of re-synchronizing the entire network again only synchronizes the particular region which is associated with the depleted synchronizing node.