A Markovian Approach To Modeling The Optimal Lifetime Of Multi-Hop Wireless Sensor Networks

A Markov Decision Process (MDP) framework is presented for modeling the lifetime of Multi-hop Wireless Sensor Networks (WSNs). The model applies to both non-cooperative and cooperative (CT) networks. To our knowledge, this is the first work to model the lifetime of multi-hop networks that jointly considers the dynamics of MAC layer link admission, routing layer queuing and energy evolution. We propose a new algorithm that exploits the Stochastic Shortest Path (SSP) structure and Mixed Integer Linear Programming (MILP) to efficiently solve the problem. Numerical results on the optimal lifetime of non-CT and CT networks are presented to validate the model.