Information rates over an AWGN channel with an energy harvesting transmitter having a finite energy buffer

We consider an AWGN channel with a transmitter powered by an energy harvesting source. The node is equipped with a finite energy buffer. Such a system can be modelled as a channel with side information (about energy in the energy buffer) causally known at the transmitter. The receiver is assumed to have no side information. First, we provide an achievable scheme that approximates the system capacity at low buffer lengths. Next we consider a scheme viz. the truncated Gaussian signalling that provides excellent approximation to system capacity at large buffer lengths. These achievable rates are easy to compute and converge to the infinite buffer capacity as the buffer length increases. The achievable schemes we discuss belong to a general class of policies called Markov energy management policies. Markov energy management policies together with Shannon strategies provide good lower bounds to the system capacity.