Machine Learning for Model Order Selection in MIMO OFDM Systems

TL;DR

This paper introduces a machine learning approach for accurately estimating the number of multipath components in MIMO OFDM channels, outperforming existing methods especially in challenging environments with closely spaced MPCs.

Contribution

The paper presents a novel ML-based method leveraging multidimensional features of MIMO OFDM systems for improved model order selection in complex scenarios.

Findings

Higher accuracy than state-of-the-art methods in nearly coherent environments

Enhanced reliability in MPC estimation

Effective in environments with closely spaced multipath components

Abstract

A variety of wireless channel estimation methods, e.g., MUSIC and ESPRIT, rely on prior knowledge of the model order. Therefore, it is important to correctly estimate the number of multipath components (MPCs) which compose such channels. However, environments with many scatterers may generate MPCs which are closely spaced. This clustering of MPCs in addition to noise makes the model order selection task difficult in practice to currently known algorithms. In this paper, we exploit the multidimensional characteristics of MIMO orthogonal frequency division multiplexing (OFDM) systems and propose a machine learning (ML) method capable of determining the number of MPCs with a higher accuracy than state of the art methods in almost coherent scenarios. Moreover, our results show that our proposed ML method has an enhanced reliability.