Machine Learning to Enhance the V2V Communication-based IoD Using Companding Dataset

This paper investigates the improvement of Vehicle-to-Vehicle (V2V) communications in the Internet of Drones (IoD) scenario for the reduction of the Peak-to-Average Power Ratio (PAPR) problem in Orthogonal Frequency Division Multiplexing (OFDM) systems. Motivated by the recent research on machine learning and deep neural network, this study will attempt to optimize Quality of Service (QoS) by means of a multiprocessing approach based on Daubechies wavelets. The proposed scheme integrates high-performance PAPR reduction schemes, such as µ-law companding, hybrid precoding-companding, and neural network-based adaptive algorithms. Deep reinforcement learning-driven approach in clipping and filtering is further proposed, which can offer real-time adaptation for the high-demanding communications. Results show a considerable gain in PAPR, over 60% PAPR reduction under high SNR conditions. These findings show that IoD can sustain stable V2V connections in urban use cases, thereby providing a scalable solution for 5G and beyond networks.