Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Cluster-wise processing in fronthaul-aware cell-free massive MIMO systems

Zahra Mobini, Ahmet Hasim Gokceoglu, Li Wang, Gunnar Peters, Hyundong Shin, Hien Quoc Ngo

TL;DR

This work addresses scalability in fronthaul-limited CF-mMIMO by proposing cluster-wise processing, where APs are grouped into processing clusters with local CSI to reduce signaling and computation. It introduces a hybrid SLINR-based pseudo-SE metric and develops two modified WMMSE algorithms to jointly optimize precoding, power, and user association at different time scales (instantaneous and statistical CSI-based). The results show that cluster-wise processing with an appropriate number of PCs achieves near network-wide performance while significantly reducing fronthaul requirements, with statistical CSI-based designs offering favorable complexity-performance trade-offs in large-scale deployments. The findings highlight the practical viability of cluster-wise CF-mMIMO and provide design guidance on when to deploy instantaneous versus statistical CSI-based optimization under realistic fronthaul constraints.

Abstract

We exploit a general cluster-based network architecture for a fronthaul-limited user-centric cell-free massive multiple-input multiple-output (CF-mMIMO) system under different degrees of cooperation among the access points (APs) to achieve scalable implementation. In particular, we consider a CF-mMIMO system wherein the available APs are grouped into multiple processing clusters (PCs) to share channel state information (CSI), ensuring that they have knowledge of the CSI for all users assigned to the given cluster for the purposes of designing resource allocation and precoding. We utilize the sum pseudo-SE metric, which accounts for intra-cluster interference and intercluster-leakage, providing a close approximation to the true sum achievable SE. For a given PC, we formulate two optimization problems to maximize the cluster-wise weighted sum pseudo-SE under fronthaul constraints, relying solely on local CSI. These optimization problems are associated with different computational complexity requirements. The first optimization problem jointly designs precoding, user association, and power allocation, and is performed at the small-scale fading time scale. The second optimization problem optimizes user association and power allocation at the large-scale fading time scale. Accordingly, we develop a novel application of modified weighted minimum mean square error (WMMSE)-based approach to solve the challenging formulated non-convex mixed-integer problems.

Cluster-wise processing in fronthaul-aware cell-free massive MIMO systems

TL;DR

This work addresses scalability in fronthaul-limited CF-mMIMO by proposing cluster-wise processing, where APs are grouped into processing clusters with local CSI to reduce signaling and computation. It introduces a hybrid SLINR-based pseudo-SE metric and develops two modified WMMSE algorithms to jointly optimize precoding, power, and user association at different time scales (instantaneous and statistical CSI-based). The results show that cluster-wise processing with an appropriate number of PCs achieves near network-wide performance while significantly reducing fronthaul requirements, with statistical CSI-based designs offering favorable complexity-performance trade-offs in large-scale deployments. The findings highlight the practical viability of cluster-wise CF-mMIMO and provide design guidance on when to deploy instantaneous versus statistical CSI-based optimization under realistic fronthaul constraints.

Abstract

We exploit a general cluster-based network architecture for a fronthaul-limited user-centric cell-free massive multiple-input multiple-output (CF-mMIMO) system under different degrees of cooperation among the access points (APs) to achieve scalable implementation. In particular, we consider a CF-mMIMO system wherein the available APs are grouped into multiple processing clusters (PCs) to share channel state information (CSI), ensuring that they have knowledge of the CSI for all users assigned to the given cluster for the purposes of designing resource allocation and precoding. We utilize the sum pseudo-SE metric, which accounts for intra-cluster interference and intercluster-leakage, providing a close approximation to the true sum achievable SE. For a given PC, we formulate two optimization problems to maximize the cluster-wise weighted sum pseudo-SE under fronthaul constraints, relying solely on local CSI. These optimization problems are associated with different computational complexity requirements. The first optimization problem jointly designs precoding, user association, and power allocation, and is performed at the small-scale fading time scale. The second optimization problem optimizes user association and power allocation at the large-scale fading time scale. Accordingly, we develop a novel application of modified weighted minimum mean square error (WMMSE)-based approach to solve the challenging formulated non-convex mixed-integer problems.
Paper Structure (23 sections, 1 theorem, 50 equations, 10 figures, 2 tables, 2 algorithms)

This paper contains 23 sections, 1 theorem, 50 equations, 10 figures, 2 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Key Contributions
  3. Generic Cluster-Wise Processing-Based Network Architecture
  4. Uplink Training for Channel Estimation
  5. Downlink Payload Data Transmission
  6. Fronthaul Requirements
  7. Instantaneous CSI-Based Cluster-Wise Sum pseudo-SE Maximization
  8. Hybrid Leakage-Intra-Cluster-Interference
  9. Problem Formulation
  10. Modified WMMSE-Based Approach
  11. Statistical CSI-Based Cluster-Wise Sum Hardening-Based Pseudo-SE Maximization
  12. Problem Formulation and Modified WMMSE-Based Solution
  13. Numerical Results
  14. Parameters and Setup
  15. Results and Discussions
  16. ...and 8 more sections

Key Result

Proposition 1

The weighted sum-pseudo-SE maximization problem eq:problem5 has the same solution as the following WMMSE problem: where $\rho_{k}$ represents the MSE weight for user $k$ and $e_{k}$ shows the corresponding MSE, which is given by

Figures (10)

  • Figure 1: User centric CF-mMIMO with cluster-wise processing.
  • Figure 2: Instantaneous CSI-based design Algorithm 1
  • Figure 3: Statistical CSI-based design Algorithm 2
  • Figure 5: Comparison among the sum-SE achieved by different optimization approaches, where $L=24$, $K=15$, $M=10$, $\mathrm{FH_{max}}=10$ Gbps, and $M_{\mathrm{mo}}=32$.
  • Figure 6: Average sum SE versus number of AP antennas, $L$, where $K=15$, $M=8$, $\mathrm{FH_{max}}=10$ Gbps, and $M_{\mathrm{mo}}=32$.
  • ...and 5 more figures

Theorems & Definitions (4)

  • Remark 1
  • Remark 2
  • Proposition 1
  • proof