Methods for Simplifying Quasi-Deterministic Millimeter-Wave Channel Models

Abstract

It is crucial to have a dependable and precise channel model in order to study the properties of millimeter wave (mmWave) propagation. The Quasi-deterministic (QD) channel model is employed in this viewpoint, which describes the propagation of mm-Wave as a group of reflected and scattered rays originating from a complex environmental setup. These rays are assumed to travel in clusters, with each cluster consisting of a deterministic ray followed by postcursor rays and preceded by precursor rays. The summation of these rays is the number of multiple path components (MPCs) of each cluster. However, this comes at the cost of higher computational complexity for the channel model, which can hinder the simulation’s scalability. To simplify the QD channel model while maintaining accuracy, one option is to decrease the number of MPCs. In this paper, we present an analysis of path gains (PGs) of specular and diffused rays to reduce the total number of MPCs. Specifically, we propose two different reduction methods namely: i) the reduced post rays (RPR) method ii) the removed surfaces post rays (RSPR) method. The computational performance of the proposed methods is investigated in terms of computational time, and complexity. Additionally, the accuracy validation compared to the original QD model is evaluated in terms of PG cumulative distribution function (CDF), signal-to-noise ratio (SNR), and intra-cluster statistics. The proposed methods’ complexity and accuracy were assessed by examining measured data from indoor and outdoor environments at 60 GHz and 28 GHz, respectively. Both first and second-order reflection orders were tested to illustrate the balance between the two variables. The simplified methods suggested can decrease computational time by approximately 16% and 11% for RSPR and RPR schemes, respectively, when compared to the original QD.