On the Relay-Fallback Tradeoff in Millimeter Wave Wireless System

TL;DR

This paper compares fallback and relay options in millimeter wave systems, analyzing their performance based on blockage, traffic, and network parameters to determine optimal choices.

Contribution

It introduces a novel blockage model and provides a criterion to select the best option between fallback and relay in mmWave networks.

Findings

Fallback is optimal for light traffic and low blockage probability.

Relay is preferable for heavy traffic and semi-static topology.

Performance depends on payload size, beam training overhead, and blockage probability.

Abstract

Millimeter wave (mmWave) communications systems are promising candidate to support extremely high data rate services in future wireless networks. MmWave communications exhibit high penetration loss (blockage) and require directional transmissions to compensate for severe channel attenuations and for high noise powers. When blockage occurs, there are at least two simple prominent options: 1) switching to the conventional microwave frequencies (fallback option) and 2) using an alternative non-blocked path (relay option). However, currently it is not clear under which conditions and network parameters one option is better than the other. To investigate the performance of the two options, this paper proposes a novel blockage model that allows deriving maximum achievable throughput and delay performance of both options. A simple criterion to decide which option should be taken under which network condition is provided. By a comprehensive performance analysis, it is shown that the right option depends on the payload size, beam training overhead, and blockage probability. For a network with light traffic and low probability of blockage in the direct link, the fallback option is throughput- and delay-optimal. For a network with heavy traffic demands and semi-static topology (low beam-training overhead), the relay option is preferable.