Evolution of NOMA Toward Next Generation Multiple Access (NGMA)

Yuanwei Liu, Shuowen Zhang, Xidong Mu, Zhiguo Ding, Robert Schober, Naofal Al-Dhahir, Ekram Hossain, Xuemin Shen

Introduction

As the wireless industry pivots toward 6G, the limitations of current Orthogonal Multiple Access (OMA) schemes, such as the OFDMA used in 5G, are becoming apparent. OMA allocates dedicated resource blocks—be they time, frequency, or code—to individual users. While simple, this approach is increasingly inefficient for the "Internet of Everything" (IoE), which demands massive connectivity, Terabit-per-second data rates, and sub-millisecond latency.

Non-Orthogonal Multiple Access (NOMA) emerged as a solution by allowing multiple users to share the same resource blocks in the power domain. However, to meet 6G's aggressive targets, NOMA must evolve. This paper proposes a transition to "Next Generation Multiple Access" (NGMA), a unified framework that generalizes NOMA and Space-Division Multiple Access (SDMA) to handle the complexities of future networks.

The NOMA Foundation

NOMA is more than just a spectral efficiency booster; in Single-Input Single-Output (SISO) systems, it is theoretically optimal for achieving the capacity region. Unlike OMA, which is strictly suboptimal, NOMA utilizes Successive Interference Cancellation (SIC) to decode signals from multiple users sharing the same spectrum.

Comparison of NOMA and OMA principles, including power-domain multiplexing and rate regions.

As shown in the figure above, NOMA's advantage is most pronounced in "overloaded" scenarios—where the number of users exceeds the available orthogonal resources or antennas. By leveraging the power domain, NOMA can support a higher density of connections, a critical requirement for 6G.

A Unified Framework for NGMA

The core contribution of this research is a proposed unified multi-antenna framework for both downlink and uplink communications. This framework bridges the gap between NOMA and SDMA (beamforming).

  • Downlink: Users are organized into clusters. Within each cluster, NOMA (via SIC) is employed to manage interference between users with different channel conditions. Across different clusters, SDMA beamforming is used to isolate signals.
  • Uplink: The system employs a layered detection approach. Within a specific layer, parallel detection (SDMA) ensures high-speed processing. Across different layers, serial detection (NOMA/SIC) is used to cancel inter-layer interference.
The proposed unified NGMA framework illustrating (a) single-cluster NOMA and (b) multi-cluster beamforming-based NOMA.

This unified approach allows the network to flexibly adjust its operation along a spectrum between pure SDMA and pure NOMA, depending on real-time channel conditions and user requirements.

Synergies with 6G Technologies

NGMA does not exist in a vacuum; its performance is amplified when integrated with other 6G candidates:

  • Reconfigurable Intelligent Surfaces (RIS): RIS can "shape" wireless channels to artificially create the channel gain disparities that NOMA thrives on.
  • Integrated Sensing and Communication (ISaC): NOMA can facilitate the simultaneous transmission of sensing and communication signals over the same resources.
  • Machine Learning: Modern tools like Deep Reinforcement Learning and Federated Learning are essential for managing the increased optimization complexity of NGMA systems.

Key Results and Future Challenges

Simulations demonstrate that NGMA significantly outperforms traditional OMA. In robotic communication scenarios, NOMA-enabled access points provided a high-rate coverage area more than twice as large as OMA. Furthermore, in RIS-aided systems, the spectral efficiency gap between NOMA and OMA widens as the number of reflecting elements increases.

Despite these gains, several hurdles remain:

  1. SIC Error Propagation: If the first signal in a sequence is decoded incorrectly, errors propagate through the entire chain.
  2. CSI Requirements: NGMA requires highly accurate Channel State Information (CSI), which is difficult to maintain in high-mobility scenarios (e.g., UAVs) or high-frequency bands (THz).
  3. Complexity vs. Latency: The serial nature of SIC can introduce latency, necessitating a careful balance to meet 6G's ultra-low latency goals.

In conclusion, the evolution from NOMA to NGMA represents a fundamental shift toward a more flexible, capacity-achieving multiple access paradigm, setting the stage for the next decade of wireless innovation.

Summary updated

BibTeX
@article{liu2022evolution,
  author = {Yuanwei Liu and Shuowen Zhang and Xidong Mu and Zhiguo Ding and Robert Schober and Naofal Al-Dhahir and Ekram Hossain and Xuemin Shen},
  title = {{Evolution of NOMA Toward Next Generation Multiple Access (NGMA)}},
  journal = {IEEE J. Sel. Areas Commun.},
  year = {2022},
  doi = {10.1109/jsac.2022.3145234},
}
Figure