Zero-forcing Oriented Power Minimization for Multi-cell MISO-NOMA Systems: A Joint User Grouping, Beamforming and Power Control Perspective

Yaru Fu, Mingshan Zhang, Lou Salaün, Chi Wan Sung, Chung Shue Chen

TL;DR

This paper tackles the challenge of minimizing power consumption in multi-cell MISO-NOMA wireless systems by proposing a novel, distributed approach that jointly optimizes user grouping, beamforming, and power allocation. The proposed methods significantly outperform existing schemes in terms of power efficiency, outage rate, and energy efficiency, paving the way for more sustainable future wireless networks.

What Problem Does This Paper Address?

Future wireless communication systems face immense pressure to increase data rates and connectivity while simultaneously reducing power consumption due to operator profitability and global carbon emission concerns. Multi-cell Multiple-Input Single-Output Non-Orthogonal Multiple Access (MISO-NOMA) systems offer a promising solution for achieving high spectrum efficiency by allowing multiple users to share the same resource block. However, optimizing these systems is complex, involving a mixed integer non-convex programming problem that is difficult to solve directly. Specifically, efficiently grouping users, designing beamforming vectors, and allocating power across multiple cells to minimize total system power, while meeting individual user data rate requirements and respecting constraints like maximum users per cluster, remains a significant challenge.

Methods

The authors propose a two-fold methodology to tackle the complex power minimization problem:

  1. Joint User Grouping and Beamforming (BF):
    • Zero-Forcing Beamforming (ZF-BF): This near-optimal and computationally efficient BF technique is employed.
  2. Semiorthogonal User Selection (SUS): An iterative algorithm is used to select a subset of "BF users" in each cell, which are crucial for generating the ZF-BF vectors.
  3. User Clustering: Two novel user clustering strategies are introduced:
  4. Channel Condition based User Clustering (CCUC): Users are grouped based solely on their channel conditions, offering low computational complexity and independent optimization from power control.
  5. Power Consumption based User Clustering (PCUC): This strategy considers both channel conditions and inter-cell interference to minimize each cell's power consumption. It is optimized jointly with power control and formulated as a linear sum assignment problem.
  6. Optimal Power Allocation:
    • After user grouping and BF are determined, the problem is transformed into a non-convex power control problem.
  7. The feasibility and optimality of this problem are analyzed, and an iterative algorithm is developed to find the optimal power allocation. This algorithm is proven to converge to a unique optimal solution.

The entire approach is designed to be distributed, meaning each cell only requires local information (link gains, estimated inter-cell interference) to operate.

Datasets

Not discussed in the paper. The evaluation is based on Monte-Carlo simulations with parameters defined in Table I.

Key Insights and Ideas

  • Joint Optimization is Crucial: The paper highlights that effectively minimizing power in MISO-NOMA systems requires a holistic approach that jointly considers user grouping, beamforming, and power allocation, rather than optimizing them in isolation.
  • PCUC's Advantage: The Power Consumption based User Clustering (PCUC) strategy significantly outperforms the Channel Condition based User Clustering (CCUC) by incorporating inter-cell interference information, leading to better power savings and reduced outage rates.
  • Distributed Approach is Feasible and Effective: The proposed distributed algorithms, relying only on local information, demonstrate strong performance, making them practical for real-world deployment.
  • ZF-BF as a Practical Choice: While not strictly optimal, ZF-BF is shown to be a highly effective and practical choice for MISO-NOMA systems, balancing performance with computational feasibility.
  • Decoding Order Matters: The paper proves that a specific decoding order based on normalized interference plus noise values is optimal for minimizing power within a cluster.

Key Results

  • Power Consumption Reduction: MISO-NOMA-PCUC demonstrates substantial power savings, achieving approximately 47% and 94% reduction compared to MISO-NOMA-CCUC and MISO-OMA-ZF, respectively. These gains are more pronounced with a larger number of users per cell.
  • Reduced Outage Rate: The proposed MISO-NOMA strategies, particularly PCUC, significantly reduce system and user outage rates compared to conventional OMA-based MISO schemes and the non-clustered MISO-NOMA approach.
  • Improved Energy and Connectivity Efficiency: MISO-NOMA-PCUC exhibits the highest energy efficiency and connectivity efficiency among all evaluated schemes, crucial metrics for future wireless networks.
  • Convergence Guarantee: The proposed iterative power allocation algorithm is mathematically proven to converge to a unique optimal solution.

Limitations and Future Work

  • SUS Factor Optimization: The optimal value of the SUS factor 'a' is determined through empirical simulations. While the paper suggests further exhaustive search could improve optimality, it remains a parameter that requires careful tuning.
  • Practical Implementation of SIC: The paper assumes perfect successive interference cancellation (SIC). In real-world scenarios, imperfect SIC can degrade performance, which is not explicitly addressed.
  • Future Work: The authors suggest that future research could explore NOMA and multi-antenna techniques for other advanced applications like mobile edge computing, fog radio access, mission-critical IoT, and massive machine type communications in B5G systems.

Summary updated

Abstract

International audience
BibTeX
@inproceedings{fu2020zero,
  author = {Yaru Fu and Mingshan Zhang and Lou Salaün and Chi Wan Sung and Chung Shue Chen},
  title = {{Zero-forcing Oriented Power Minimization for Multi-cell MISO-NOMA Systems: A Joint User Grouping, Beamforming and Power Control Perspective}},
  booktitle = {HAL},
  year = {2020},
}
Figure