Energy-efficient scheduling of delay constrained traffic under practical power model

In this paper, we focus on the delay constrained scheduling problem for wireless link communication under the practical concerns, such as the practical power consumption of transmitter and energy used for channel estimation. In particular, we target to design an energy-efficient scheduling policy that minimizes the total energy expenditure while satisfying the deadline constraint based on knowledge about the unserved bits, channel state/statistics, practical power model and the remaining time slots to the deadline. We formulate the scheduling problem as a stochastic programming problem and propose a suboptimal policy by analyzing the structure of the solution. In addition, we study a special case that the whole packet is served in a single slot and propose an optimal channel-threshold policy based on the stopping theory. Numerical results illustrate that the proposed policies provide great energy efficiency especially when the data amount is relatively small and the transmission independent power consumption is large.