We consider a node in a multihop wireless network that is responsible for transmitting messages in a timely manner while being prudent about energy consumption. The node uses energy harvesting in the sense that it is powered by batteries that are charged by renewable energy sources such as wind or solar. To strike a balance between latency and availability, we develop a multi-timescale model. At a faster timescale, the node makes decisions based on local information such as queue lengths of packets in input buffers and available energy levels. The decisions include scheduling packets on the output buffers that would be transmitted at the next opportunity, possibly using network coding. At the slower timescale, we model the energy levels in the battery using a stochastic fluid-flow model to determine the availability and time-averaged latency. Ultimately, we present a unified framework that iteratively sets model parameters to satisfy latency and availability targets. The methods are based upon Markov decision processes, Markov chains, and semi-Markov process analysis.

We consider a relay node that stochastically receives packets from two opposing flows. Whenever opportunities exist, the relay performs network coding to efficiently transmit packets. However, on one hand, because of the stochastic nature, as well as possible asymmetry between the opposing flows, it would not be possible to always code packets. On the other hand, waiting for a coding opportunity could result in excessive latency, and one may be better off transmitting packets without coding. Thus, one needs to decide at each transmission opportunity whether to transmit a packet uncoded or wait for a future transmission opportunity. To enable us to optimally make that decision, we consider costs for transmission and delay, and formulate our problem as a Markov decision process. We show that the optimal policy is threshold type under a sufficient condition, and we compute it by modeling the resulting system as a Markov chain. Through numerical analysis, we show the effectiveness of the threshold policy in the relay node network, as well as in a line network scenario. Further, we compare the threshold policy against a number of simple heuristic policies and identify situations where these policies can be effective.