Studies on reliable and accurate services for event sensing, reporting, and monitoring applications.

摘要

In this dissertation, a two-tier network infrastructure, which is composed of an upper-tier service network and multiple lower-tier wireless sensor networks (WSNs), is proposed to support event sensing, reporting, and monitoring applications based on unreliable multi-hop wireless networks. Three functional modules are studied to solve critical problems in this infrastructure.In order to improve the reliability of reporting and monitoring services for sensed events in the upper-tier service network, a Self-Nomination Aided Routing Protocol (SNARP) is proposed to handle packet losses caused by transmission failures, link failures, and network congestion simultaneously. The joint sender-receiver decision-making mechanism and the timeout triggered retransmission mechanism are integrated into a store-and-forwarding packet delivery process to improve the routing reliability. Simulation results show that SNARP can recover more than 99% of lost packets when the packet loss rate is 20% at each hop. It consumes 30% less energy than Extremely Opportunistic Routing (ExOR) while achieving similar packet recovery performance.A distributed node positioning protocol, Protocol for Location And Coordination Estimation (PLACE), is proposed to estimate positions of sensing nodes within the lower-tier WSNs. With the assistance of incomplete information, PLACE first calculates the positions of nodes and then refines the positioning results according to additional references. By solving the practical issues of the alias problem and the range-measurement error problem, the node positioning accuracy of PLACE is 10 times higher than that of DV-Distance in dense WSNs.Time-synchronization is essential in providing useful timestamps in the lower-tier WSNs. An analytical model is proposed in this dissertation to analyze the accuracy of time synchronization. To the best knowledge of the author, it is the first analytical model that studies the effects of processing delay, accessing delay, propagation delay, and receiving delay on the accuracy of time synchronization approaches. The proposed analytical model can help application designers in time synchronization approach selection, performance improvement, and parameter tuning according to the requirements of target applications.