International Journal of Advanced Innovative Technology in Engineering (IJAITE)

---

OFDM Networks for Enhanced BER And PAPR Mitigation in 5G Systems

Komal Marotrao Chavhan, Prof. Dr. Ravindra M. Deshmukh

Abstract :

O-OFDMNet, an intensity-modulated direct detection transmission system, employs deep learning-aided optical orthogonal frequency division multiplexing (O-OFDM). O-OFDMNet utilizes deep neural networks (DNNs) to convert a complex-valued signal at the transmitter into a non-negative signal in the time domain and vice versa at the receiver. Unlike traditional radio frequency (RF) OFDM, O-OFDMNet retains the related frequency-domain signal processing. Unlike current O-OFDM schemes which rely on the Hermitian symmetry of the spectral-domain signal to ensure the real-valuedness of the time-domain signal, our approach achieves equivalent spectral efficiency to the RF scheme, a milestone previously unattained by any existing methods. As an autoencoder architecture, the example shows that O-OFDMNet may be taught in an end-to-end manner to simultaneously improve the bit errors ratios and the transmissions peak-to-average ratio of power. across frequency-selective channels and additive white Gaussian noise environments. Additionally, technique achieves throughput comparable to RF-OFDM, significantly surpassing that of traditional O-OFDM.

Keywords :

Full Text :

Download PDF

DOI :

Cite this paper :

References :

---