Design of Fault-Tolerant Structures for Underwater Sensor Networks Based on Markov Chains.

Abstract

Recently, underwater sensor networks (USN) have gained widespread adoption, proving instrumental in diverse research endeavors within the marine environment and across various scientific and technological domains. Considering the challenging operational conditions, a paramount issue in the present day is the creation of USNs with a predetermined fault tolerance margin. The primary focus of this study is the development and evaluation of a design method for fault-tolerant structures of USNs using Markov chains. The proposed method enables the creation of suitable USNs' structures, ensuring a guaranteed attainment of the desired number of operational cycles before failure at reasonable costs. Utilizing the mathematical framework of Markov chains, coupled with the strategy of a well-selected sequence of hierarchy levels, enables the precise determination of the network's actual number of operation cycles before failure as well as performing efficient reservation without excessive redundancy. To evaluate the efficacy of the proposed method, this study undertakes a specific case - the development of a fault-tolerant structure for a real USN with predetermined parameters. The analysis of the results obtained confirms the high feasibility of utilizing the developed method for crafting USNs for various purposes, ensuring a predefined level of fault tolerance while maintaining acceptable costs.