A Framework of Mahalanobis-Distance Metric with Supervised Learning for Clustering Multipath Components in MIMO Channel Analysis

TL;DR

This paper introduces a Mahalanobis-distance metric framework with supervised learning for improved clustering of multipath components in MIMO channel analysis, enhancing clustering quality without extensive parameter tuning.

Contribution

It proposes a general Mahalanobis-distance metric framework, unifies existing metrics, and introduces machine learning algorithms for supervised metric learning in MPC clustering.

Findings

Significantly improves clustering quality of MPCs.

The proposed metric reduces the need for parameter tuning.

Learning MPC labels requires limited effort.

Abstract

As multipath components (MPCs) are experimentally observed to appear in clusters, cluster-based channel models have been focused in the wireless channel study. However, most of the MPC clustering algorithms for MIMO channels with delay and angle information of MPCs are based on the distance metric that quantifies the similarity of two MPCs and determines the preferred cluster shape, greatly impacting MPC clustering quality. In this paper, a general framework of Mahalanobis-distance metric is proposed for MPC clustering in MIMO channel analysis, without user-specified parameters. Remarkably, the popular multipath component distance (MCD) is proved to be a special case of the proposed distance metric framework. Furthermore, two machine learning algorithms, namely, weak-supervised Mahalanobis metric for clustering and supervised large margin nearest neighbor, are introduced to learn the distance metric. To evaluate the effectiveness, a modified channel model is proposed based on the 3GPP spatial channel model to generate clustered MPCs with delay and angular information, since the original 3GPP spatial channel model (SCM) is incapable to evaluate clustering quality. Experiment results show that the proposed distance metric can significantly improve the clustering quality of existing clustering algorithms, while the learning phase requires considerably limited efforts of labeling MPCs.