Energy-efficient packet transmission over a wireless link

TLDR

Reducing transmission power and code rate can lower packet energy, but delay‑sensitive data limits how long transmissions can be extended. The study aims to minimize energy for packet transmission over a wireless link by designing lazy schedules that vary transmission times while respecting deadlines or delay constraints. We derive an optimal offline schedule under a deadline constraint and analytically extend it to the infinite‑deadline case, yielding a probabilistic characterization that informs a lazy online algorithm. Simulations demonstrate that the proposed lazy online schedule, which adapts transmission times to backlog, achieves near‑optimal energy efficiency and outperforms a deterministic fixed‑time schedule while maintaining queue stability.

Abstract

The paper considers the problem of minimizing the energy used to transmit packets over a wireless link via lazy schedules that judiciously vary packet transmission times. The problem is motivated by the following observation. With many channel coding schemes, the energy required to transmit a packet can be significantly reduced by lowering transmission power and code rate and therefore transmitting the packet over a longer period of time. However, information is often time-critical or delay-sensitive and transmission times cannot be made arbitrarily long. We therefore consider packet transmission schedules that minimize energy subject to a deadline or a delay constraint. Specifically, we obtain an optimal offline schedule for a node operating under a deadline constraint. An inspection of the form of this schedule naturally leads us to an online schedule which is shown, through simulations, to perform closely to the optimal offline schedule. Taking the deadline to infinity, we provide an exact probabilistic analysis of our offline scheduling algorithm. The results of this analysis enable us to devise a lazy online algorithm that varies transmission times according to backlog. We show that this lazy schedule is significantly more energy-efficient compared to a deterministic (fixed transmission time) schedule that guarantees queue stability for the same range of arrival rates.