DSpace Arşivi :: by Yazar "Naser, Aktham Hikmat Naser" değerine göre listeleniyor

Listeleniyor 1 - 1 / 1

Optimum ber in non-orthogonal multiple access for 5G wireless technology
(İstanbul Gelişim Üniversitesi Lisansüstü Eğitim Enstitüsü, 2024) Naser, Aktham Hikmat Naser
Non-orthogonal multiple access (NOMA) is a promising technology that has the potential to deliver high spectral efficiency, reduced latency, improved reliability, and substantial connectivity. This primarily focuses on achieving optimal bit error rate (BER) performance. It utilizes recurrent neural networks (RNN) to efficiently decode messages for multiple users, treating each time step as a separate subcarrier. A deep neural network (DNN) is employed to eliminate the need for explicit channel estimation, leveraging its ability to autonomously learn patterns in data. This proves advantageous in tasks like decoding messages in traditional NOMA systems, simplifying the receiver's operation, and mitigating error propagation effects. Through careful simulation, the BER of the system is evaluated over varying signal-to-noise ratios (SNR) in additive white Gaussian noise (AWGN) and Rayleigh distribution systems. The results show that the DNN exhibits superior BER performance compared to least square (LS), maximum likelihood (ML), and minimum mean square error (MMSE). Utilizing a QPSK demodulation scheme with wide SNR range, the DNN exhibits superior BER performance compared to LS, ML, and MMSE across all symbol combinations for two users and for two tested cases of subcarrier pilot numbers with 16 & 64.