A time-inhomogeneous Markov chain and its distributed solution for message dissemination in OUSNs

Abstract

Opportunistic underwater sensor networks (OUSNs) are deployed for various underwater applications, such as underwater creatures tracking and tactical surveillance. The index of freshness-cost ratio is firstly introduced to describe the objectives of spatio-temporal cost minimization and message freshness preservation for the message dissemination in OUSNs. To analyze the message propagation process, a time-inhomogeneous ergodic Markov chain is constructed, and the delivery progress of each message is determined by the number of hops from the destination to the nearest message holder, which is mapped into several propagation states of the Markov chain. Then, a message dissemination method (MDM) derived from a Markov decision process is specifically designed to improve the freshness-cost ratio. In the MDM, the message holders decide how many copies should be disseminated at each time slot, according to the Markov decision process. The performance of MDM is analyzed through simulation experiments that produce preferable tradeoff results, which indicate that MDM can deliver the messages with fewer spatio-temporal cost while preserving the message freshness as much as possible.