Energy and Bursty Packet Loss Tradeoff over Fading Channels: A System Level Model

TL;DR

This paper presents a multiuser scheduling scheme for 5G fading channels that balances energy efficiency and packet loss, considering imperfect channel information and application-specific QoS requirements.

Contribution

It introduces a cross-layer opportunistic scheduling framework that accounts for channel estimation errors and packet loss parameters, optimizing energy and QoS trade-offs.

Findings

Packet loss parameters are tightly coupled.

Relaxing one loss parameter offers limited benefits if others are constrained.

Channel uncertainty significantly impacts scheduler design.

Abstract

Energy efficiency and quality of service (QoS) guarantees are the key design goals for the 5G wireless communication systems. In this context, we discuss a multiuser scheduling scheme over fading channels for loss tolerant applications. The loss tolerance of the application is characterized in terms of different parameters that contribute to quality of experience for the application. The mobile users are scheduled opportunistically such that a minimum QoS is guaranteed. We propose an opportunistic scheduling scheme and address the cross layer design framework when channel state information is not perfectly available at the transmitter and the receiver. We characterize the system energy as a function of different QoS and channel state estimation error parameters. The optimization problem is formulated using Markov chain framework and solved using stochastic optimization techniques. The results demonstrate that the parameters characterizing the packet loss are tightly coupled and relaxation of one parameter does not benefit the system much if the other constraints are tight. We evaluate the energy-performance trade-off numerically and show the effect of channel uncertainty on the packet scheduler design.