Cooperative Relay Scheduling in Energy Harvesting Sensor Networks

Abstract

This chapter studies relay scheduling in wireless sensor networks with energy harvesting and cooperative communications capabilities. Sensor networks are increasingly deployed in inaccessible and remote regions for applications such as environmental monitoring, relief operations, and defense. In such networks, energy harvesting and cooperative communication paradigms are used simultaneously to design energy-efficient relay scheduling strategies. The following scheduling problem is formulated to maximize the network utility: Given an estimate of the current network state, should a source sensor node transmit its data directly to the destination sensor node, or should it use a relay sensor node to help with the transmission? An upper bound is obtained on the performance of an arbitrary scheduler. Scheduling policies are then developed to choose the appropriate transmission mode depending upon the available energy at the sensors as well as the states of their energy harvesting and data generation processes. Two separate scenarios are considered where the state of the relay node is either fully or partially observable at the source node, and the scenarios are modeled using a Markov Decision Process (MDP) and a Partially Observable Markov Decision Process (POMDP) respectively. It is shown that the POMDP can be transformed into an equivalent MDP. Optimal scheduling strategies are evaluated using value iteration algorithm, and various insights towards optimal relay scheduling are discussed. Simulation results are used to show the performance of the proposed strategies.