SWIPT Using Hybrid ARQ Over Time Varying Channels

We consider a class of wireless powered devices employing hybrid automatic repeat request to ensure reliable end-to-end communications over a two-state time-varying channel. A receiver, with no power source, relies on the energy transferred by a simultaneous wireless information and power transfer enabled transmitter to receive and decode information. Under the two-state channel model, information is received at two different rates while it is only possible to harvest energy in one of the states. The receiver aims to decode its messages with minimum expected number of re-transmissions. Dynamic and continuous nature of the problem motivated us to use a novel Markovian framework to bypass the complexities plaguing the conventional approaches such as Markov decision process. Using the theory of absorbing Markov chains, we show that there exists an optimal policy utilizing the incoming RF signal solely to harvest energy or to accumulate mutual information. Hence, we convert the original problem with continuous action and state space into an equivalent one with discrete state and action space. For independent and identically distributed channels, we prove the optimality of a simple-to-implement harvest-first-store-later type policy. However, for time-correlated channels, we demonstrate that statistical knowledge of the channel may significantly improve the performance over such policies.