Kumar Arun, Nanthaamornphong Aziz, Masud Mehedi
Department of Electronics and Communication Engineering, Sikkim Manipal Institute of Technology, Sikkim Manipal University, Majitar, Rangpo, India.
College of Computing, Prince of Songkla University, Phuket, Thailand.
Sci Rep. 2025 Aug 23;15(1):31022. doi: 10.1038/s41598-025-16414-6.
Non-Orthogonal Multiple Access (NOMA) has emerged as a prominent technique for enhancing spectral efficiency in beyond-fifth-generation (5G) and sixth-generation (6G) wireless systems. However, its performance is highly dependent on channel conditions, necessitating robust spectrum-sensing methods. This study proposes a novel Recurrent Neural Network (RNN)-based Bidirectional Long Short-Term Memory (RNN-Bi-LSTM) model to enhance the spectral performance of NOMA under various channel conditions, particularly the Rician and Rayleigh fading channels. The proposed approach was evaluated using key performance metrics, including the probability of detection (PD), probability of false alarm (PFA), bit error rate (BER), and power spectral density (PSD). Simulation results show that RNN-Bi-LSTM achieves 100% PD at - 5 dB and - 2.5 dB SNR, outperforming conventional methods such as RNN (- 3 dB and - 1.5 dB), LSTM (- 1 dB and 0.3 dB), CSD (0.2 dB and 5.5 dB), matched filter (MF) (1 dB and 0.5 dB), and energy detection (ED) (2.3 dB and 2.7 dB) in Rician and Rayleigh channels, respectively. Additionally, the RNN-Bi-LSTM model shows a 23.36% improvement in PSD suppression under Rician conditions compared with Rayleigh conditions, reflecting the benefits of LoS-enhanced propagation in reducing spectral leakage and improving detection accuracy. BER performance also improves, achieving 10⁻ at 8.8 dB and 5.8 dB SNR, whereas other methods require higher SNR. Furthermore, the model provides a more accurate PSD estimation, reduces spectral leakage, and enhances the spectrum utilization. Overall, RNN-Bi-LSTM demonstrated superior adaptability to varying channel conditions, making it a robust and efficient solution for NOMA-based spectrum sensing in advanced wireless communication systems.
非正交多址接入（NOMA）已成为一种在超第五代（5G）和第六代（6G）无线系统中提高频谱效率的重要技术。然而，其性能高度依赖于信道条件，因此需要强大的频谱感知方法。本研究提出了一种基于递归神经网络（RNN）的双向长短期记忆（RNN-Bi-LSTM）模型，以增强NOMA在各种信道条件下的频谱性能，特别是在莱斯和瑞利衰落信道中。使用包括检测概率（PD）、误报概率（PFA）、误码率（BER）和功率谱密度（PSD）在内的关键性能指标对所提出的方法进行了评估。仿真结果表明，RNN-Bi-LSTM在信噪比为-5 dB和-2.5 dB时实现了100%的检测概率，在莱斯和瑞利信道中分别优于传统方法，如RNN（-3 dB和-1.5 dB）、LSTM（-1 dB和0.3 dB）、CSD（0.2 dB和5.5 dB）、匹配滤波器（MF）（1 dB和0.5 dB）和能量检测（ED）（2.3 dB和2.7 dB）。此外，与瑞利条件相比，RNN-Bi-LSTM模型在莱斯条件下的PSD抑制提高了23.36%，这反映了视距增强传播在减少频谱泄漏和提高检测精度方面的优势。BER性能也有所提高 在信噪比为8.8 dB和5.8 dB时达到10⁻，而其他方法需要更高的信噪比。此外，该模型提供了更准确的PSD估计，减少了频谱泄漏，并提高了频谱利用率。总体而言，RNN-Bi-LSTM对不同信道条件表现出卓越的适应性，使其成为先进无线通信系统中基于NOMA的频谱感知的强大而高效的解决方案。
