Low-Dimensionality of Noise-Free RSS and its Application in Distributed Massive MIMO

TL;DR

This paper reveals that noise-free uplink RSS data in distributed massive MIMO systems has a low-dimensional structure, which is exploited by a novel reconstruction-based Gaussian process method to improve user location prediction accuracy.

Contribution

The paper introduces RecGP, a new method that leverages low-dimensionality of noise-free RSS for enhanced location prediction in massive MIMO systems.

Findings

RecGP outperforms standard GP methods in prediction accuracy.

Low-dimensional principal subspace effectively captures noise-free RSS characteristics.

Reconstruction-based approach reduces prediction error in noisy test scenarios.

Abstract

We examine the dimensionality of noise-free uplink received signal strength (RSS) data in a distributed multiuser massive multiple-input multiple-output system. Specifically, we apply principal component analysis to the noise-free uplink RSS and observe that it has a low-dimensional principal subspace. We make use of this unique property to propose RecGP - a reconstruction-based Gaussian process regression (GP) method which predicts user locations from uplink RSS data. Considering noise-free RSS for training and noisy test RSS for location prediction, RecGP reconstructs the noisy test RSS from a low- dimensional principal subspace of the noise-free training RSS. The reconstructed RSS is input to a trained GP model for location prediction. Noise reduction facilitated by the reconstruction step allows RecGP to achieve lower prediction error than standard GP methods which directly use the test RSS for location prediction.