Date of Completion

5-9-2014

Embargo Period

5-7-2014

Major Advisor

Dr. Reda Ammar

Co-Major Advisor

Dr. Jun-Hong Cui

Associate Advisor

Sanguthevar Rajasekaran

Associate Advisor

Bing Wang

Field of Study

Computer Science and Engineering

Degree

Doctor of Philosophy

Open Access

Campus Access

Abstract

Underwater Sensor Network (UWSN) is the enabling technology for a broad range of potential and evolving applications in scientific, industrial and military domains. To make these applications viable, acoustic waves are used as an alternative to the highly attenuated electromagnetic waves in water medium. However, acoustic channel introduces unique challenges that are absent in UWSNs counterpart, terrestrial Wireless Sensor Networks (WSNs). In this dissertation we tackle these challenges mainly by designing routing, single sink and multi-gateway redeployment strategies. Firstly, we have developed an Adaptive Power Controlled Routing (APCR) strategy that is location free, energy efficient and is robust to the mobility of underwater nodes. In our techniques, nodes assign themselves to concentric layers centered at the sink node and the routing paths are the determined on the fly. Nodes can adjust their transmission power to a finite set of values to adjust for the change in the network topology to maintain connectivity and reduce the energy consumption Secondly, we have opted to adaptively redeploy the surface sink to reduce the effect of mobility on the network performance metrics. In our adaptive dynamic sink redeployment strategy, the sink moves only if an actual reduction in energy consumption is expected. The redeployment decision is based on the routing information collected at the surface sink throughout the network's operation. The surface sink finds and moves to the optimal new location that minimizes total energy consumption. Lastly, we have extended our redeployment strategy for multi-surface gateway architecture. In which underwater nodes communicate with one of the surface gateways that collectively form a virtual sink. For multi-surface gateway redeployment, we use APCR as the underlying routing protocol to determine when to redeploy. The redeployment problem is then reduced to find the optimal node to gateway assignment. The solution minimizes the total number of transmission power levels used by nodes communicating with surface gateways while enforcing a set of constraints on both surface gateways and underwater nodes. We solved for the optimal solution using exhaustive search and compared it with a near optimal solution obtained by a greedy algorithm to cater for processing limitation of surface gateways.

Download
COinS