PRNet with Convolution Layer for PAPR Reduction of OFDM Signals and FPGA Implementation

This research employs deep learning to address the high peak-to-average power ratio (PAPR) in orthogonal frequency division multiplexing (OFDM), which is critical for wireless communications. Although the PAPR-reducing network (PRNet), a deep learning model, effectively suppresses PAPR, it requires substantial computational resources. In this research, the number of layers in a PRNet model is optimized and a fully connected layer can be replaced with a convolution layer to reduce the computational load. Furthermore, the FPGA system is designed using a hardware description language, and the effectiveness of the proposed method is confirmed through an estimation of resource utilization.