AI Insight
This study introduces the Graph-Guided Adaptive Companding (GGAC) framework, a novel signal processing method designed to reduce Peak-to-Average Power Ratio (PAPR) in Power-Domain Non-Orthogonal Multiple Access (PD-NOMA) systems combined with OFDM multicarrier modulation. By modeling the composite signal as a graph where user components are represented as nodes with weighted edges encoding power and interference relationships, the framework applies node-specific companding parameters that selectively suppress peak-dominant signals while protecting low-power and SIC-critical users. Simulation results indicate PAPR reductions of up to 10 dB over conventional PD-NOMA, alongside improvements in bit error rate, SINR, and a reduction in out-of-band radiation exceeding 40 dB.
Why it matters
High PAPR in 5G and beyond systems reduces power amplifier efficiency and increases hardware costs and energy consumption, so an effective reduction technique could meaningfully improve the viability and energy efficiency of dense wireless networks. This framework has direct practical implications for the design of base stations and user devices in future 5G and 6G deployments serving large numbers of simultaneous users.
by Arun Kumar, Mehedi Masud, Mansor Alohali, Prashanta Chandra Pradhan, Aziz Nanthaamornphong
Power-domain Non-Orthogonal Multiple Access (PD-NOMA) is a promising multiple-access technique for fifth-generation (5G) and beyond-fifth-generation (B5G/6G) wireless networks because of its ability to improve spectral efficiency, massive connectivity, and user fairness through superposition coding and successive interference cancellation (SIC). However, when combined with multicarrier modulation such as Orthogonal Frequency Division Multiplexing (OFDM), PD-NOMA suffers from a high Peak-to-Average Power Ratio (PAPR), which degrades the power amplifier efficiency, increases nonlinear distortion, and significantly impairs the bit error rate (BER) and SIC reliability. Existing PAPR reduction techniques, particularly conventional companding schemes, apply uniform or globally adaptive nonlinear transformations and fail to account for the inherent multiuser coupling and SIC sensitivity of PD-NOMA signals. To address these limitations, this study proposes a novel Graph-Guided Adaptive Companding (GGAC) framework for PD-NOMA systems. The proposed method models the composite PD-NOMA waveform as a graph, where the signal components are represented as nodes, and their power and interference relationships are captured through weighted edges. Graph-based importance metrics are then used to assign node-specific companding parameters, enabling the selective suppression of peak-dominant components while preserving the integrity of low-power and SIC-critical users. The simulation results demonstrate that GGAC achieves a significant PAPR reduction, offering gains of up to 10 dB over conventional PD-NOMA. In addition, the proposed framework significantly improves the BER and Signal-to-Interference-plus-Noise Ratio (SINR), reduces out-of-band radiation by more than 40 dB, and preserves the constellation geometry with a lower error vector magnitude. These results confirm that GGAC provides an effective, low-complexity, and scalable solution for enhancing the energy efficiency and reliability in future PD-NOMA-based 5G systems.
Source: Graph-guided adaptive companding for PAPR reduction in power-domain NOMA systems