Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Deep-Learning-Based Channel Estimation for Distributed MIMO with 1-bit Radio-Over-Fiber Fronthaul

Alireza Bordbar, Lise Aabel, Christian Häger, Christian Fager, Giuseppe Durisi

TL;DR

The paper addresses uplink channel estimation in distributed MIMO systems using 1-bit radio-over-fiber fronthaul, where the fronthaul conveys two-level quantized signals. It adapts a deep-unfolding maximum-likelihood estimator from prior work to handle oversampling, dithering, and the impairments introduced by AGCs and comparators, and extends the model to reflect practical nonidealities. Results show the proposed DL estimator significantly outperforms the Bussgang-based BLMMSE method and exhibits robustness to the added distortions, with gains persisting under additive noise and with refined training data. The work highlights a viable path toward accurate reciprocity-based coherent downlink beamforming in low-resolution fronthaul networks and motivates real-world over-the-air validation on the referenced testbed.

Abstract

We consider the problem of pilot-aided, uplink channel estimation in a distributed massive multiple-input multiple-output (MIMO) architecture, in which the access points are connected to a central processing unit via fiber-optical fronthaul links, carrying a two-level-quantized version of the received analog radio-frequency signal. We adapt to this architecture the deep-learning-based channel-estimation algorithm recently proposed by Nguyen et al. (2023), and explore its robustness to the additional signal distortions (beyond 1-bit quantization) introduced in the considered architecture by the automatic gain controllers (AGCs) and by the comparators. These components are used at the access points to generate the two-level analog waveform from the received signal. Via simulation results, we illustrate that the proposed channel-estimation method outperforms significantly the Bussgang linear minimum mean-square error channel estimator, and it is robust against the additional impairments introduced by the AGCs and the comparators.

Deep-Learning-Based Channel Estimation for Distributed MIMO with 1-bit Radio-Over-Fiber Fronthaul

TL;DR

The paper addresses uplink channel estimation in distributed MIMO systems using 1-bit radio-over-fiber fronthaul, where the fronthaul conveys two-level quantized signals. It adapts a deep-unfolding maximum-likelihood estimator from prior work to handle oversampling, dithering, and the impairments introduced by AGCs and comparators, and extends the model to reflect practical nonidealities. Results show the proposed DL estimator significantly outperforms the Bussgang-based BLMMSE method and exhibits robustness to the added distortions, with gains persisting under additive noise and with refined training data. The work highlights a viable path toward accurate reciprocity-based coherent downlink beamforming in low-resolution fronthaul networks and motivates real-world over-the-air validation on the referenced testbed.

Abstract

We consider the problem of pilot-aided, uplink channel estimation in a distributed massive multiple-input multiple-output (MIMO) architecture, in which the access points are connected to a central processing unit via fiber-optical fronthaul links, carrying a two-level-quantized version of the received analog radio-frequency signal. We adapt to this architecture the deep-learning-based channel-estimation algorithm recently proposed by Nguyen et al. (2023), and explore its robustness to the additional signal distortions (beyond 1-bit quantization) introduced in the considered architecture by the automatic gain controllers (AGCs) and by the comparators. These components are used at the access points to generate the two-level analog waveform from the received signal. Via simulation results, we illustrate that the proposed channel-estimation method outperforms significantly the Bussgang linear minimum mean-square error channel estimator, and it is robust against the additional impairments introduced by the AGCs and the comparators.
Paper Structure (10 sections, 12 equations, 3 figures, 1 table)

This paper contains 10 sections, 12 equations, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. System Model
  3. Distributed MIMO with 1-Bit Radio-Over-Fiber Fronthaul
  4. Pilot-Aided Channel Estimation: a Mathematical Model
  5. Deep-Learning-Based Channel Estimation
  6. Maximum-Likelihood Channel Estimation
  7. Structure of the
  8. Refined System Model
  9. Simulation Results
  10. Conclusions

Figures (3)

  • Figure 1: Block diagram of one and the in the distributed architecture with 1-bit radio-over-fiber fronthaul considered in the paper.
  • Figure 2: as a function of signal-to-dither ratio $E_{\text{s}}\xspace / E_{\text{d}}\xspace$ for Bussgang and -based channel estimators.
  • Figure 3: The effect of and ADC bit flips on the performance of the proposed -based channel estimator.