Genetic Algorithms for solving Coverage Problem in 3D Underwater Wireless Sensor Networks

Abstract

Underwater Wireless Sensor Networks (UWSNs) are playing as an important tool in the exploration and monitoring of underwater areas which are not easily reachable by human being. Under water sensor nodes are mostly used in the applications, e.g., oceanographic data collection, pollution monitoring, offshore exploration, disaster prevention, assisted navigation and tactical surveillance. UWSNs mainly consists of inexpensive nodes equipped with memory, processors and short rang wireless communication, each capable of collecting, storing and processing environmental information over wide geographic areas and communicating with neighboring nodes through wireless links.

Major challenges in the design and development of UWSNs are: more energy consumption, limited bandwidth and memory power, propagation delay, sensor density, link redundancy, sensor deployment cost, high bit error rate, temporary losses of connectivity. Some specific characteristics of UWSNs such as random deployment in 3D environment, self organization, mobility of sensors and underwater obstacles make terrestrial WSN protocols and services become infeasible and more challenging in underwater. Among the fundamental services, coverage refers to how well a sensing field is monitored by sensors which measures the Quality of Service (QoS) of the network. In the literature, coverage problem has been proven to be NP-complete.

Our objective is to maximize the sensing coverage considering 3D deployment and mobility with minimum energy consumption. Two Genetic Algorithm (GA) based solutions for the 3D coverage problem has been suggested. The first approach assumes static node deployment and the second approach assumes mobility along z-axis of the 3D deployment. Performance and proposed solutions has been evaluated with inhouse simulator for different performance metric such as coverage rate, energy consumption, scalability and mobility.