Energy efficient transmission scheduling over Mars proximity links

We consider the scheduling of transmissions from an energy limited Mars lander to a Mars orbiter. Typically, the channel quality of the Mars proximity links is time-varying due to orbital dynamics, multi-path effects and the antenna positioning on the lander. Since the channel state determines the throughput obtained per unit of energy transmitted, it is desirable to select when, and at what data rate, to transmit based on channel conditions. In this paper we consider the dual problem of maximizing the throughput of a lander that has a limited amount of energy to be used for transmission; and minimizing the energy consumption used for transmission of data subject to delay constraints. In (Tarello et al., 2005) it was shown, using techniques from dynamic programming, that energy efficiency can be significantly improved through an adaptive rate and power allocation scheme. We apply the policies developed in (Tarello et al., 2005) to the Mars proximity link and compare them with scheduling algorithms presently in use for proximity links. Our simulations, using channel measurement data obtained from NASA's 2001 Mars Odyssey orbiter and the Mars Exploration Rover (MER) project, show over an order of magnitude increase in throughput and decrease in energy consumption through the use of the dynamic programming based rate and power adaptation scheme