A hybrid deep learning and companding technique for distortion-resilient optical NOMA VLC with 1024-QAM modulation
-
Murigendrayya M. Hiremath
Abstract
Visible light communication (VLC) has gained significant attention as a next-generation wireless technology due to its unlicensed bandwidth, high security, and seamless integration with LED illumination. To support high data rates and user scalability, power-domain non-orthogonal multiple access (NOMA) with 1024-QAM modulation enhances spectral efficiency; however, this combination introduces high peak-to-average power ratio (PAPR), resulting in nonlinear signal distortion and degraded bit error rate (BER) performance in intensity-modulated optical systems. To overcome these limitations, this paper proposes a deep learning-based nonlinear companding framework for waveform optimization in Optical NOMA. The proposed technique significantly reduces PAPR, achieving 3.2 dB, 4.4 dB, and 5.3 dB at a CCDF of 10−3 for 256, 512, and 1024 subcarriers, respectively, outperforming A-Law, μ-Law, and PTS by up to 7.5 dB improvement. BER analysis further confirms its superiority, achieving a BER of 10−3 at only 8.9 dB SNR, compared to 15.2 dB (A-Law), 13.1 dB (μ-Law), 11.4 dB (PTS), and 17.8 dB for the unprocessed Optical NOMA waveform. These results demonstrate improved power efficiency, reduced nonlinear distortion, and enhanced detection reliability, making the proposed method a strong candidate for high-speed VLC-based NOMA systems.
-
Research ethics: Not applicable.
-
Informed consent: Not applicable.
-
Author contributions: The authors have accepted responsibility for the entire content of this manuscript and approved its submission.
-
Use of Large Language Models, AI and Machine Learning Tools: Not applicable.
-
Conflict of interest: The authors state no conflict of interest.
-
Research funding: Not applicable.
-
Data availability: Not applicable.
References
1. Kumar, A, Gaur, N, Nanthaamornphong, A. Reducing PAPR in NOMA waveforms using genetic-enhanced PTS and SLM: a low-complexity approach for improved throughput, power spectral density, and power efficiency. Results Eng 2025;26:104893. https://doi.org/10.1016/j.rineng.2025.104893.Suche in Google Scholar
2. Al-Hashmi, T, Tarhuni, N, Mesbah, M, Asif, H. PAPR reduction for OFDM-Over-NOMA using hybrid partial selective mapping and artificial neural networks. In: 2025 international conference for artificial intelligence, applications, innovation and ethics (AI2E). Oman, Muscat; 2025. p. 1–6.10.1109/AI2E64943.2025.10983531Suche in Google Scholar
3. Kumar, A, Gour, N, Sharma, H, Pareek, R. A hybrid technique for the PAPR reduction of NOMA waveform. Int J Commun Syst 2023;36. https://doi.org/10.1002/dac.5412.Suche in Google Scholar
4. Sharan, N, Ghorai, SK, Kumar, A. PAPR reduction using a precoder and compander combination in a NOMA-OFDM VLC system. In: 2022 2nd international conference on artificial intelligence and signal processing (AISP). Vijayawada, India; 2022. p. 1–4.10.1109/AISP53593.2022.9760659Suche in Google Scholar
5. Tan, J, Wang, Q, Wang, Z. Modified PTS-based PAPR reduction for ACO-OFDM in visible light communications. Sci China Inf Sci 2015;58:1–3. https://doi.org/10.1007/s11432-015-5414-7.Suche in Google Scholar
6. Taha, TB, Fayed, HA, Aly, MH, Mahmoud, M. A reduced PAPR hybrid OFDM visible light communication system. Opt Quant Electron 2022;54. https://doi.org/10.1007/s11082-022-04219-0.Suche in Google Scholar
7. Miriyala, G, Mani, VV. A new PAPR reduction technique in DCO-OFDM for visible light communication systems. Opt Commun 2020;474:126064. https://doi.org/10.1016/j.optcom.2020.126064.Suche in Google Scholar
8. Shatti, AS, Sabri, SS. PAPR reduction using VLM precoding with nonlinear companding for DCO-OFDM based VLC systems. Int J Electr Comput Eng 2020;10:714–23.Suche in Google Scholar
9. Khalaf, H, Al-Halafi, A, Ragab, AM, Alsharef, M. Mixed nonlinear companding transforms for PAPR reduction in ACO-OFDM VLC systems. Opt Commun 2021;499:1–9.Suche in Google Scholar
10. Ramazan, M, Khan, B, Zaman, F. μ-law nonlinear companding with modified region for PAPR reduction in ACO-OFDM VLC systems. Opt Laser Technol 2021;145:1–9.Suche in Google Scholar
11. Azarnia, R, Zand, P, Afsahi, AM. Clipping noise mitigation using compressive sensing for optical OFDM systems. IEEE Photon Technol Lett 2018;30:273–6.Suche in Google Scholar
12. Mounir, M, Youssef, MI, Aboshosha, AM. Low-complexity selective mapping technique for PAPR reduction in downlink power-domain OFDM-NOMA. EURASIP J Appl Signal Process 2023;2023:1–13. https://doi.org/10.1186/s13634-022-00968-y.Suche in Google Scholar
13. Sayyari, S, Saeedi-Sourck, H, Mohammadi, M. An efficient PAPR reduction scheme for OFDM-NOMA systems. Int J Commun Syst 2021;34:1–14.Suche in Google Scholar
14. George, NA, Mishra, SK. PAPR reduction in F-NOMA using selective mapping method. In: Proceedings 2023 14th international conference on computing communication and networking technologies (ICCCNT), Delhi, India; 2023. p. 1–5.10.1109/ICCCNT56998.2023.10308006Suche in Google Scholar
15. Kumar, A, Rajagopal, K, Gugapriya, G, Sharma, H, Gour, N, Masud, M, et al.. Reducing PAPR with low complexity filtered NOMA using novel algorithm. Sustainability 2022;14:9631. https://doi.org/10.3390/su14159631.Suche in Google Scholar
16. Kumar, A, Rajagopal, K, Alruwais, N, Mesfer Alshahrani, H, Mahgoub, H, Othman, KM. PAPR reduction using SLM-PTS-CT hybrid PAPR method for optical NOMA waveform. Heliyon 2023;9. https://doi.org/10.1016/j.heliyon.2023.e20901.Suche in Google Scholar PubMed PubMed Central
17. Kumar, A. A low complex PTS-SLM-companding technique for PAPR reduction in 5G NOMA waveform. Multimed Tool Appl 2024;83:45141–62. https://doi.org/10.1007/s11042-023-17223-7.Suche in Google Scholar
© 2025 Walter de Gruyter GmbH, Berlin/Boston
