On the Delay-Throughput Tradeoff in Distributed Wireless Networks

TL;DR

This paper analyzes the delay-throughput tradeoff in a single-hop wireless network with fading channels, proposing a new effective throughput metric, and characterizes conditions for zero packet loss while maximizing throughput.

Contribution

It introduces the effective throughput concept, derives its asymptotic scaling, and studies the delay and dropping probability trade-offs under various packet arrival processes.

Findings

Effective throughput scales as (log n)/α̂ for large n.

Conditions for zero packet dropping probability are established.

Trade-offs between throughput, delay, and dropping probability are characterized.

Abstract

This paper deals with the delay-throughput analysis of a single-hop wireless network with $n$ transmitter/receiver pairs. All channels are assumed to be block Rayleigh fading with shadowing, described by parameters $(\alpha,\varpi)$, where $\alpha$ denotes the probability of shadowing and $\varpi$ represents the average cross-link gains. The analysis relies on the distributed on-off power allocation strategy (i.e., links with a direct channel gain above a certain threshold transmit at full power and the rest remain silent) for the deterministic and stochastic packet arrival processes. It is also assumed that each transmitter has a buffer size of one packet and dropping occurs once a packet arrives in the buffer while the previous packet has not been served. In the first part of the paper, we define a new notion of performance in the network, called effective throughput, which captures the effect of arrival process in the network throughput, and maximize it for different cases of packet arrival process. It is proved that the effective throughput of the network asymptotically scales as $\frac{\log n}{\hat{\alpha}}$, with $\hat{\alpha} \triangleq \alpha \varpi$, regardless of the packet arrival process. In the second part of the paper, we present the delay characteristics of the underlying network in terms of the packet dropping probability. We derive the sufficient conditions in the asymptotic case of $n \to \infty$ such that the packet dropping probability tend to zero, while achieving the maximum effective throughput of the network. Finally, we study the trade-off between the effective throughput, delay, and packet dropping probability of the network for different packet arrival processes.