Interference Mitigation via Rate-Splitting and Common Message Decoding in Cloud Radio Access Networks

Alaa Alameer Ahmad, Hayssam Dahrouj, Anas Chaaban, Aydin Sezgin, Mohamed-Slim Alouini

Abstract

Cloud-radio access networks (C-RAN) help overcoming the scarcity of radio resources by enabling dense deployment of base-stations (BSs), and connecting them to a central-processor (CP). This paper considers the downlink of a C-RAN, where the cloud is connected to the BSs via limited-capacity backhaul links. The paper proposes splitting the message of each user into two parts, a private part decodable at the intended user only, and a common part which can be decoded at a subset of users, as a means to enable large-scale interference management in CRAN. To this end, the paper optimizes a transmission scheme that combines rate splitting (RS), common message decoding (CMD), clustering and coordinated beamforming. The paper focuses on maximizing the weighted sum-rate subject to per-BS backhaul capacity and transmit power constraints, so as to jointly determine the RS-CMD mode of transmission, the cluster of BSs serving private and common messages of each user, and the associated beamforming vectors of each user private and common messages. The paper proposes solving such a complicated non-convex optimization problem using $l_0$-norm relaxation techniques, followed by inner-convex approximations (ICA), so as to achieve stationary solutions to the relaxed non-convex problem. Numerical results show that the proposed method provides significant performance gain as compared to conventional interference mitigation techniques in CRAN which treat interference as noise (TIN).

Introduction

In modern wireless networks, the dense deployment of base stations (BSs) in Cloud Radio Access Networks (C-RAN) is essential to meet skyrocketing data demands. However, this proximity introduces two critical bottlenecks: severe inter-cell interference and limited backhaul capacity between the BSs and the central processor. Traditional strategies, such as "Treating Interference as Noise" (TIN), are often inefficient when interference is strong. This paper proposes a novel framework that moves beyond TIN by utilizing Rate-Splitting (RS) and Common Message Decoding (CMD) to actively manage interference.

Proposed Methodology

The core of the proposed scheme is the decomposition of user data into "private" and "common" components. As shown in the system model below, the central processor splits each user's message $v_k$ into a private part $v_k^p$ (decodable only by the target user) and a common part $v_k^c$ (decodable by a group of users).

System model of a C-RAN architecture employing Rate-Splitting and Common Message Decoding.

These messages are encoded into symbols $s_k^p$ and $s_k^c$ and distributed across BSs via backhaul links ($C_1, C_2, C_3$). Users employ Successive Interference Cancellation (SIC) to decode the common messages first, subtract that interference, and then decode their own private messages.

The authors formulate a joint optimization problem to maximize the Weighted Sum-Rate (WSR). This involves dynamic clustering (deciding which BSs serve which messages) and coordinated beamforming. To solve this NP-hard Mixed-Integer Non-Linear Problem (MINLP), the authors utilize $l_0$-norm relaxation to handle discrete clustering choices and Inner Convex Approximation (ICA) to iteratively solve the non-convex problem.

Simulation Setup

The performance was evaluated using numerical simulations of a 7-cell wrapped-around network with 28 users.

  • Channel Models: The simulation incorporated Rayleigh small-scale fading, 8 dB log-normal shadowing, and a standard path-loss model ($140.7 + 36.7 \log_{10}(d)$).
  • Hardware: Each BS was equipped with 8 antennas, operating on a 10 MHz bandwidth.
  • Heuristics: A percentile-based heuristic was used to identify "weak" users who would benefit most from decoding the common messages of "strong" interferers.

Key Results and Discussion

The proposed RS-CMD scheme demonstrates substantial improvements over conventional TIN-based systems:

  • Throughput Gains: In scenarios with a 200m inter-cell distance, RS-CMD outperformed static TIN by 42.3%. When the distance increased to 400m, the gain jumped to 61.11%.
  • Backhaul Efficiency: RS-CMD makes significantly better use of limited backhaul, reaching performance plateaus much earlier than TIN. This suggests that RS-CMD is a vital tool for navigating the middle ground between infinite and zero backhaul scenarios.
  • Interference Management: By turning a portion of interference into decodable common information, the scheme effectively clears the airwaves for private messages, particularly in high-power transmission regimes where interference is most disruptive.

Limitations and Future Work

While effective, the RS-CMD approach increases computational complexity due to the iterative ICA algorithm and the requirement for SIC at the user terminal. Furthermore, the current model assumes perfect Channel State Information (CSI), which may not be available in real-world deployments. Future research could explore machine learning to optimize common message sets and investigate the impact of imperfect CSI on the system's performance.

Summary updated

Abstract

Cloud-radio access networks (C-RAN) help overcoming the scarcity of radio resources by enabling dense deployment of base-stations (BSs), and connecting them to a central-processor (CP). This paper considers the downlink of a C-RAN, where the cloud is connected to the BSs via limited-capacity backhaul links. The paper proposes splitting the message of each user into two parts, a private part decodable at the intended user only, and a common part which can be decoded at a subset of users, as a means to enable large-scale interference management in CRAN. To this end, the paper optimizes a transmission scheme that combines rate splitting (RS), common message decoding (CMD), clustering and coordinated beamforming. The paper focuses on maximizing the weighted sum-rate subject to per-BS backhaul capacity and transmit power constraints, so as to jointly determine the RS-CMD mode of transmission, the cluster of BSs serving private and common messages of each user, and the associated beamforming vectors of each user private and common messages. The paper proposes solving such a complicated non-convex optimization problem using $l_0$-norm relaxation techniques, followed by inner-convex approximations (ICA), so as to achieve stationary solutions to the relaxed non-convex problem. Numerical results show that the proposed method provides significant performance gain as compared to conventional interference mitigation techniques in CRAN which treat interference as noise (TIN).
BibTeX
@inproceedings{ahmad2019interference,
  author = {Alaa Alameer Ahmad and Hayssam Dahrouj and Anas Chaaban and Aydin Sezgin and Mohamed-Slim Alouini},
  title = {{Interference Mitigation via Rate-Splitting and Common Message Decoding in Cloud Radio Access Networks}},
  booktitle = {IEEE Access},
  year = {2019},
  url = {https://openreview.net/forum?id=g6IhypF6X3m},
}
Figure