Energy-Efficient Data Transmission with A Non-FIFO Packet

TL;DR

This paper addresses energy-efficient data transmission with non-FIFO packets by transforming the problem into an optimal split factor search, proposing an algorithm and an online policy that outperform FIFO-based methods.

Contribution

It introduces a novel packet split and reorder process for non-FIFO packets and develops an optimal algorithm using existing strategies, enhancing energy efficiency in transmission.

Findings

Proposed an optimal offline transmission scheme for non-FIFO packets.

Developed an online policy with performance comparable to the offline scheme.

Proved the optimality of the algorithm in energy-efficient transmission.

Abstract

This paper investigates the problem of energy-efficient packet transmission with a non-FIFO Packet over a point-to-point additive white Gaussian noise (AWGN) time-invariant channel under the feasibility constraints. More specifically, we consider the scenario where there is a packet that has a deadline that is earlier than that of the previously arrived packet. For this problem, the First-In-First-Out (FIFO) transmission mode adopted in the existing literatures is no longer optimal. We first propose a novel packet split and reorder process which convert the inconsistency in the order of deadlines and arrival instants of the packet sequence into a consistent one. After the split and reorder process, the original problem considered in this paper is transformed into the problem of finding the optimal split factor. We propose an algorithm that finds the split factor which consists of checking four possibilities by applying the existing optimal transmission strategy \emph{"String Tautening"} for FIFO packets. In addition, we prove the optimality of the proposed algorithm in the presence of a non-FIFO packet by exploiting the optimality properties of the most energy efficient transmission strategy. Based on the proposed optimal offline scheme, an efficient online policy which assumes causal arrival information is also studied and shown to achieve a comparable performance to the proposed optimal offline scheme.