User Association and Interference Management in Massive MIMO HetNets

Qiaoyang Ye, Ozgun Y. Bursalioglu, Haralabos C. Papadopoulos, Constantine Caramanis, Jeffrey G. Andrews

Abstract

Two key traits of 5G cellular networks are much higher base station (BS) densities - especially in the case of low-power BSs - and the use of massive MIMO at these BSs. This paper explores how massive MIMO can be used to jointly maximize the offloading gains and minimize the interference challenges arising from adding small cells. We consider two interference management approaches: joint transmission (JT) with local precoding, where users are served simultaneously by multiple BSs without requiring channel state information exchanges among cooperating BSs, and resource blanking, where some macro BS resources are left blank to reduce the interference in the small cell downlink. A key advantage offered by massive MIMO is channel hardening, which enables to predict instantaneous rates a priori. This allows us to develop a unified framework, where resource allocation is cast as a network utility maximization (NUM) problem, and to demonstrate large gains in cell-edge rates based on the NUM solution. We propose an efficient dual subgradient based algorithm, which converges towards the NUM solution. A scheduling scheme is also proposed to approach the NUM solution. Simulations illustrate more than 2x rate gain for 10th percentile users vs. an optimal association without interference management.

Introduction

The evolution toward 5G networks is characterized by two primary trends: a significant increase in base station (BS) density through the deployment of low-power small cells and the integration of massive MIMO technology. While these "HetNets" (Heterogeneous Networks) promise massive capacity gains, they introduce complex challenges in user association and interference management. Traditionally, users connect to the BS with the strongest signal, often leading to overloaded macro cells and underutilized small cells. Furthermore, offloading users to small cells exposes them to severe interference from nearby macro BSs.

Methodology

This paper proposes a unified framework to jointly optimize offloading and interference management using a Network Utility Maximization (NUM) approach. The framework leverages two primary strategies:

Local Joint Transmission (LJT)

LJT is a form of Coordinated Multi-Point (CoMP) where multiple BSs simultaneously serve a single user. Unlike traditional joint transmission which requires extensive backhaul communication for Channel State Information (CSI) exchange, LJT utilizes only locally obtained CSI at each BS. This significantly reduces overhead while still providing the diversity gains needed for cell-edge users.

Figure 1: Illustration of Local Joint Transmission (LJT) where multiple base stations serve users using local precoding without inter-BS CSI exchange.

Resource Blanking and Channel Hardening

To mitigate macro-to-small-cell interference, the framework employs resource blanking, where specific macro BS resources are muted to create "quiet zones." A critical enabler for this optimization is the "channel hardening" property of Massive MIMO. As the number of antennas grows, fast-fading fluctuations average out, making instantaneous data rates highly predictable. This predictability allows the decoupling of long-term user association from short-term packet scheduling.

Figure 3: Contrast scenario showing Joint Precoding, which requires high-overhead CSI exchange between base stations, a challenge LJT seeks to avoid.

Proposed Algorithms

The authors develop a dual subgradient-based algorithm to solve the NUM problem efficiently. This algorithm finds the optimal user association and resource partitioning, typically converging in fewer than 60 iterations. To bridge the gap between theoretical optimization and real-time operation, a Virtual Queue (VQ) based scheduler is introduced to map the optimal association into practical transmission slots.

Results and Discussion

System-level simulations using 3GPP benchmarks demonstrate significant performance improvements. The proposed framework achieves a 2.2x rate gain for 10th percentile (cell-edge) users compared to optimal association without interference management. Furthermore, the practical VQ scheduler achieves approximately 90% of the theoretical maximum utility, proving the framework's viability for real-world 5G deployments.

Conclusion

By exploiting the stability provided by Massive MIMO's channel hardening, this research provides a scalable solution for the HetNet interference problem. The combination of Local Joint Transmission and resource blanking, managed via a NUM framework, offers a path to significantly improved fairness and throughput for the most disadvantaged users in the network.

Summary updated

Abstract

Two key traits of 5G cellular networks are much higher base station (BS) densities - especially in the case of low-power BSs - and the use of massive MIMO at these BSs. This paper explores how massive MIMO can be used to jointly maximize the offloading gains and minimize the interference challenges arising from adding small cells. We consider two interference management approaches: joint transmission (JT) with local precoding, where users are served simultaneously by multiple BSs without requiring channel state information exchanges among cooperating BSs, and resource blanking, where some macro BS resources are left blank to reduce the interference in the small cell downlink. A key advantage offered by massive MIMO is channel hardening, which enables to predict instantaneous rates a priori. This allows us to develop a unified framework, where resource allocation is cast as a network utility maximization (NUM) problem, and to demonstrate large gains in cell-edge rates based on the NUM solution. We propose an efficient dual subgradient based algorithm, which converges towards the NUM solution. A scheduling scheme is also proposed to approach the NUM solution. Simulations illustrate more than 2x rate gain for 10th percentile users vs. an optimal association without interference management.
BibTeX
@article{ye2016user,
  author = {Qiaoyang Ye and Ozgun Y. Bursalioglu and Haralabos C. Papadopoulos and Constantine Caramanis and Jeffrey G. Andrews},
  title = {{User Association and Interference Management in Massive MIMO HetNets}},
  journal = {IEEE Trans. Communications},
  year = {2016},
  url = {https://openreview.net/forum?id=53MFbu5OwL},
}
Figure