Quasi-Deterministic Channel Model for mmWaves: Mathematical Formalization and Validation

TL;DR

This paper develops a detailed mathematical formulation for Quasi-Deterministic (QD) channel models at mmWave frequencies, validating their accuracy against real measurements and highlighting their importance for reliable 5G network design.

Contribution

It introduces a comprehensive mathematical formalization for mmWave QD channel models and validates their accuracy through comparison with real measurement data.

Findings

QD model improves KS test scores by up to 0.537 compared to deterministic models.

The model accurately characterizes multipath propagation in mmWave scenarios.

Validation against real measurements confirms the model's reliability.

Abstract

5G and beyond networks will use, for the first time ever, the millimeter wave (mmWave) spectrum for mobile communications. Accurate performance evaluation is fundamental to the design of reliable mmWave networks, with accuracy rooted in the fidelity of the channel models. At mmWaves, the model must account for the spatial characteristics of propagation since networks will employ highly directional antennas to counter the much greater pathloss. In this regard, Quasi-Deterministic (QD) models are highly accurate channel models, which characterize the propagation in terms of clusters of multipath components, given by a reflected ray and multiple diffuse components of any given Computer Aided Design (CAD) scenario. This paper introduces a detailed mathematical formulation for QD models at mmWaves, that can be used as a reference for their implementation and development. Moreover, it compares channel instances obtained with an open source NIST QD model implementation against real measurements at 60 GHz, substantiating the accuracy of the model. Results show that, when comparing the proposed model and deterministic rays alone with a measurement campaign, the Kolmogorov-Smirnov (KS) test of the QD model improves by up to 0.537.