Distributed User Clustering and Resource Allocation for Imperfect NOMA in Heterogeneous Networks

Abdulkadir Celik, Ming-Cheng Tsai, Redha M. Radaydeh, Fawaz S. Al-Qahtani, Mohamed-Slim Alouini

Abstract

In this paper, we propose a distributed cluster formation (CF) and resource allocation (RA) framework for non-ideal non-orthogonal multiple access (NOMA) schemes in heterogeneous networks. The imperfection of the underlying NOMA scheme is due to the receiver sensitivity and interference residue from non-ideal successive interference cancellation (SIC), which is generally characterized by a fractional error factor (FEF). Our analytical findings first show that several factors have a significant impact on the achievable NOMA gain. Then, we investigate fundamental limits on NOMA cluster size as a function of FEF levels, cluster bandwidth, and quality of service (QoS) demands of user equipments (UEs). Thereafter, a clustering algorithm is developed by taking feasible cluster size and channel gain disparity of UEs into account. Finally, we develop a distributed alpha-fair RA framework where alpha governs the trade-off between maximum throughput and proportional fairness objectives. Based on the derived closed-form optimal power levels, the proposed distributed solution iteratively updates bandwidths, clusters, and UEs' transmission powers. Numerical results demonstrate that proposed solutions deliver a higher spectral and energy efficiency than traditionally adopted basic NOMA cluster size of two. We also show that an imperfect NOMA cannot always provide better performance than orthogonal multiple access under certain conditions. Finally, our numerical investigations reveal that NOMA gain is maximized under downlink/uplink decoupled (DUDe) UE association.

Introduction

To meet the massive connectivity and high-throughput demands of 5G and beyond, networks are turning to Non-Orthogonal Multiple Access (NOMA). Unlike traditional Orthogonal Multiple Access (OMA) that gives each user a dedicated time or frequency slot, NOMA multiplexes multiple users onto the same frequency resource by varying their transmit power. The receiver then uses Successive Interference Cancellation (SIC) to untangle the signals.

However, most existing NOMA research relies on two unrealistic assumptions: perfect SIC and two-user clusters. In reality, hardware limits and estimation errors leave behind residual interference, quantified by a Fractional Error Factor (FEF). Limiting clusters to two users leaves potential spectral efficiency on the table, while ignoring imperfect SIC leads to power allocation strategies that fail in real-world deployments. This paper addresses these issues with a distributed, imperfection-aware approach.

Distributed Framework for Cluster Formation and Resource Allocation

The proposed framework decomposes the complex, NP-hard central optimization problem into a distributed architecture. As illustrated in Figure 1, the system consists of multiple Base Stations (BSs) that independently manage Cluster Formation (CF) and Power Allocation (PA), while a Macro Base Station (MBS) coordinates Bandwidth Allocation (BA) to ensure network-wide fairness.

Proposed distributed framework architecture showing the interaction between Base Stations (BS) for clustering and power allocation, and the Macro Base Station (MBS) for bandwidth allocation.

Methods

  1. Analytical Cluster Limits: Using random matrix theory, we derive closed-form mathematical limits for the maximum feasible cluster size based on bandwidth, user Quality of Service (QoS) demands, and the FEF.
  2. Disparity-Driven Clustering: Each BS independently groups its users to maximize channel gain disparity. We mathematically prove that the difference in signal strength between users is the primary driver of NOMA performance gains.
  3. Distributed $\alpha$-Fair Resource Allocation: The resource allocation is split into a local "Slave Problem" where clusters calculate optimal power levels using closed-form expressions that account for imperfect SIC, and a "Master Problem" where the MBS updates bandwidth allocation to achieve $\alpha$-fairness.

Impact of Imperfect SIC

A critical insight of this work is that NOMA is not a universal solution; its efficacy is highly sensitive to hardware imperfections. If the FEF is too high, residual interference can overwhelm the system, making NOMA perform worse than traditional OMA.

Optimal power weights vs. Fractional Error Factor (FEF) for different QoS demands, illustrating the transition between different operational cases as interference residue increases.

As shown in Figure 3, the optimal power weights ($\omega$) must be dynamically adjusted based on the FEF levels. The analysis identifies four distinct cases of operation. In Case 1 (low FEF), NOMA provides significant gains. However, as FEF increases (moving into Cases 2 and 3), the power allocation must shift to compensate for interference, eventually reaching a point where the system must revert to OMA-like behavior to maintain QoS.

Key Results and Conclusion

Numerical evaluations in simulated 5G environments reveal several key findings:

  • Efficiency Gains: Increasing the cluster size from 2 to 3 requires approximately 40% less total power consumption while providing a 30% increase in sum-rate, significantly boosting overall energy efficiency.
  • Fairness Scaling: Under strict fairness constraints ($\alpha \to 1$), the performance gap between the proposed large-cluster NOMA and basic 2-user NOMA widens. Basic NOMA performance can drop to 20% of the objective value, while the proposed method maintains near-optimal performance.
  • Association Strategy: NOMA gains are maximized under downlink/uplink decoupled (DUDe) UE association, which reduces interference compared to traditional coupled association.

In conclusion, while imperfect SIC presents a challenge, a distributed framework that accounts for hardware limitations and allows for larger cluster sizes can significantly outperform traditional NOMA and OMA implementations.

Summary updated

Abstract

In this paper, we propose a distributed cluster formation (CF) and resource allocation (RA) framework for non-ideal non-orthogonal multiple access (NOMA) schemes in heterogeneous networks. The imperfection of the underlying NOMA scheme is due to the receiver sensitivity and interference residue from non-ideal successive interference cancellation (SIC), which is generally characterized by a fractional error factor (FEF). Our analytical findings first show that several factors have a significant impact on the achievable NOMA gain. Then, we investigate fundamental limits on NOMA cluster size as a function of FEF levels, cluster bandwidth, and quality of service (QoS) demands of user equipments (UEs). Thereafter, a clustering algorithm is developed by taking feasible cluster size and channel gain disparity of UEs into account. Finally, we develop a distributed alpha-fair RA framework where alpha governs the trade-off between maximum throughput and proportional fairness objectives. Based on the derived closed-form optimal power levels, the proposed distributed solution iteratively updates bandwidths, clusters, and UEs' transmission powers. Numerical results demonstrate that proposed solutions deliver a higher spectral and energy efficiency than traditionally adopted basic NOMA cluster size of two. We also show that an imperfect NOMA cannot always provide better performance than orthogonal multiple access under certain conditions. Finally, our numerical investigations reveal that NOMA gain is maximized under downlink/uplink decoupled (DUDe) UE association.
BibTeX
@article{celik2019distributed,
  author = {Abdulkadir Celik and Ming-Cheng Tsai and Redha M. Radaydeh and Fawaz S. Al-Qahtani and Mohamed-Slim Alouini},
  title = {{Distributed User Clustering and Resource Allocation for Imperfect NOMA in Heterogeneous Networks}},
  journal = {IEEE Trans. Communications},
  year = {2019},
  url = {https://openreview.net/forum?id=oG7Rn78uhB},
}
Figure