An Intelligent Deep Neural Based Peak-To-Average Power Ratio Mitigated System for Multi User OFDM Channel

Abstract

An Orthogonal Frequency Division Multiplexing (OFDM) signal having a large Peak-to-Average Power ratio (PAPR) can cause noise signals and power degradation. To remove the high PAPR in OFDM signals, the proposed Strawberry-based Recurrent Neural Framework (SbRNF) model is proposed. While transmitting the signals in the OFDM channel, the rate of PAPR was high due to more subcarriers. Basically, more subcarriers are needed to convey the OFDM channel's signals. Our proposed model made the reduction of PAPR in the OFDM channel easier. The result of the proposed model (SbRNF) was determined, and signal-to-noise ratio (SNR), Throughput, and Bit Error Rate (BER) were compared with other existing methodologies. Here, the presented model is executed in the MATLAB platform and shows that SbRNF has a low Bit Error Rate (BER), high throughput value, and Cumulative Distribution Function (CCDF) performance in the presence of high PAPR. While comparing the existing models with SbRNF provides better PAPR reduction in the signal transmission.