Power-Coefficient-Aware Adaptive Companding for PAPR Reduction in Downlink PD-NOMA-OFDM Systems
Main article
Abstract
The Peak-to-Average Power Ratio (PAPR) problem presents a critical barrier to the practical deployment of Power-Domain Non-Orthogonal Multiple Access (PD-NOMA) systems when combined with Orthogonal Frequency Division Multiplexing (OFDM). Existing companding methods apply uniform or globally adaptive nonlinear transformations that overlook the intrinsic power hierarchy of PD-NOMA signals, leading to excessive distortion of low-power users and degradation of Successive Interference Cancellation (SIC) reliability. This paper proposes a novel Power-Coefficient-Aware Adaptive Companding (PCAC) framework that exploits the power allocation structure inherent in PD-NOMA superposition signals to derive user-specific companding parameters. The proposed method assigns companding strengths proportional to each user's power coefficient, thereby selectively suppressing high-amplitude peaks contributed by dominant users while preserving the integrity of weak-user signals that are most sensitive to SIC decoding failures. A closed-form SINR expression incorporating the residual companding distortion is derived, enabling analytical characterization of BER performance. Extensive MATLAB simulations under Rayleigh fading channels with QPSK and 16-QAM modulations confirm that PCAC achieves a PAPR reduction exceeding 9 dB over conventional PD-NOMA at a CCDF of 10⁻³, alongside significant improvements in BER, SINR, and spectral containment compared to fixed μ-law companding, Selective Mapping (SLM), Partial Transmit Sequences (PTS), and learning-based benchmarks. With linear computational complexity O(N), the proposed framework offers a practical, scalable, and high-performance solution for 5G and beyond-5G PD-NOMA networks.
