Centralized and Decentralized Channel Estimation in FDD Multi-User Massive MIMO Systems

Abstract

We design a centralized and a decentralized variational Bayesian learning (C-and D-VBL) algorithms for the base station (BS) of a frequency division duplex massive multiple input multiple output (mMIMO) cellular system, wherein users send compressed information for it to estimate their downlink channels. The BS in the decentralized algorithm consists of multiple processing units (PUs), and each PU separately estimates the channels of a group of users, by employing the proposed D-VBL algorithm. To reduce channel estimation error, the PUs exploit the structured sparsity inherent in multi-user mMIMO channels by exchanging information among themselves. We investigate the proposed C-VBL and low-complexity D-VBL algorithms and show that i) they substantially outperform the state-of-the-art centralized and decentralized algorithms in terms of the normalized mean squared error and the bit error rate. This is because they beneficially exploit the inherent channel sparsity, while the existing state-of-the-art solutions fail to do so. The proposed D-VBL is also robust to PU failures, and provides a similar performance as its centralized counterpart (C-VBL), but with a much reduced complexity.

Efficient Distributed Multi-Task Schemes for mmWave MIMO Channel Estimation

“…In the relevant literature, the problem is mainly addressed either in point-topoint [8], [11]- [13] or in point-to-multiple points frameworks [14]. In addition, the sparse channel recovery approaches [15] can be employed either individually [11], [16] or centrally (e.g., at the BS) [8], [17], [18]. In the centralized solutions, the CS algorithms exploit the common sparsity patterns among the channels of nearby devices that are manifested in their measurements due to the transmission environment and, thus, enhance the estimation accuracy or reduce the training overhead.…”

Limited Feedback Design Based on Kronecker Product Codebook for Massive MIMO Systems

“…To improve the accuracy of CSI in the FDD mode, a downlink training method was proposed in [5] where long-term channel statistics and previously received training signals from users are utilized. In [6], Bayesian learning-based channel estimation that enables the estimation of the downlink channel at the base station (BS) was proposed for the FDD multiuser MIMO system. In addition, partial reciprocity-based CSI acquisition was recently studied [7], [8].…”