Abstract
Utilizing mobile nodes is highly desirable for wireless sensor network deployments across a large area, possibly over a harsh or inhabitable terrain. However, their usability is doubted as they are costlier compared with static nodes, and their movement-actuation consumes a significant amount of energy. Additionally, utilization of mobile nodes imposes complex operational constraints; thesis examined the possible mechanisms for effective utilization of mobile nodes operating within the constraints. Techno-economic evaluation using geometric models concluded that mobile nodes are cost effective for use in sensor networks. It is possible to relocate mobile nodes without use of localization services, and collectively address two network attributes - coverage and clustering. A hybrid network comprising only partial mobile nodes is sufficient to fulfill coverage requirements for the network, deduced by determining the multi-parameter relations that govern node mobility. An operational mechanism that supports sensor networks to operate on undulating-harsh terrain has been developed, relying on data collection by flying sensors and bypassing the need for multi-hop communication between clusters and the base station. It was also found that seemingly non-related applications of sensor networks can be addressed collectively, utilizing passive node mobility.
Assessment Committee:
Dr. Dilip Krishnaswamy, IBM Research, USA
Dr. Periklis Chatzimisios, Alexander TEI of Thessaloniki, Greece
Dr. Rasmus Olsen (Chairman), Aalborg University, Denmark
Supervisor:
Dr. Neeli Prasad, Aalborg University, Denmark
Co-supervisor:
Dr. Rasmus Hjorth Nielsen, Aalborg University, Denmark
Moderator
Dr. Sofoklis Kyriazakos, Aalborg University, Denmark