Scalable and Accurate Modeling of the Millimeter Wave Channel

TL;DR

This paper introduces a simplified millimeter wave channel model that significantly reduces computational complexity while maintaining accuracy, enabling scalable simulations for 5G network development.

Contribution

It proposes a simplified 3GPP SCM for mmWave frequencies that reduces computation time by up to 12.5 times without sacrificing key distribution accuracy.

Findings

Reduction in computation time by up to 12.5 times

Maintains accurate distributions of SINR and other metrics

Suitable for large-scale scenario simulations

Abstract

Communication at millimeter wave (mmWave) frequencies is one of the main novelties introduced in the 5th generation (5G) of cellular networks. The opportunities and challenges associated with such high frequencies have stimulated a number of studies that rely on simulation for the evaluation of the proposed solutions. The accuracy of simulations largely depends on that of the channel model, but popular channel models for mmWaves, such as the Spatial Channel Models (SCMs), have high computational complexity and limit the scalability of the scenarios. This paper profiles the implementation of a widely-used SCM model for mmWave frequencies, and proposes a simplified version of the 3GPP SCM that reduces the computation time by up to 12.5 times while providing essentially the same distributions of several metrics, such as the Signal-to-Interference-plus-Noise Ratio (SINR) in large scale scenarios. We also give insights on the use cases in which using a simplified model can still yield valid results.