Pilot Contamination Mitigation for Wideband Massive MMO: Number of Cells Vs Multipath

TL;DR

This paper introduces a joint channel estimation and beamforming method for wideband massive MIMO systems that leverages multipath characteristics to mitigate pilot contamination, enabling multiple cells to share resources efficiently.

Contribution

It presents a novel, linear approach that exploits multipath taps to determine the maximum number of cells sharing resources while mitigating pilot contamination.

Findings

Up to 18 cells can share the same band in LTE environments.

The proposed method outperforms existing designs in simulations.

The approach is simple to implement and scalable with bandwidth.

Abstract

This paper proposes novel joint channel estimation and beamforming approach for multicell wideband massive multiple input multiple output (MIMO) systems. Using our channel estimation and beamforming approach, we determine the number of cells $N_c$ that can utilize the same time and frequency resource while mitigating the effect of pilot contamination. The proposed approach exploits the multipath characteristics of wideband channels. Specifically, when the channel has $L$ multipath taps, it is shown that $N_c\leq L$ cells can reliably estimate the channels of their user equipments (UEs) and perform beamforming while mitigating the effect of pilot contamination. For example, in a long term evolution (LTE) channel environment having delay spread $T_d=4.69\mu$ second and channel bandwidth $B=2.5$MHz, we have found that $L=18$ cells can use this band. In practice, $T_d$ is constant for a particular environment and carrier frequency, and hence $L$ increases as the bandwidth increases. The proposed channel estimation and beamforming design is linear, simple to implement and significantly outperforms the existing designs, and is validated by extensive simulations.