Energy‐efficient data transmission with non‐FIFO packets

Abstract

In this study, the authors investigate the problem of energy-efficient packet transmission with arbitrary arrival instants and deadline constraints over a point-to-point additive white Gaussian noise channel. This is different from previous work where it is assumed that the packets follow a first-in–first-out (FIFO) order in that a packet that arrives earlier has a deadline that is also earlier. They first investigate the necessary and sufficient conditions of the optimal offline transmission schedule. They then propose an algorithm which finds the transmission schedule of each packet. Next, they show that their algorithm is optimal by proving it satisfies the sufficient conditions of the optimal offline transmission schedule, and further obtain the computational complexity in the worst case: 𝒪(N 3), where N is the number of packets. In addition, based on the proposed optimal offline policy, an efficient heuristic online policy which assumes only the causal arrival information of the packets is proposed. Finally, simulation results show that the proposed offline policy can achieve significant energy savings compared with the existing FIFO offline policy, and the proposed online algorithm can achieve a comparable performance with the proposed optimal offline policy.