Predicting Downlink Retransmissions in 5G Networks Using Deep Learning

Abstract

5G networks are expected to provide high-speed, low-latency, and reliable connectivity to support various applications such as autonomous vehicles, smart cities, and the Internet of Things (IoT). However, the performance of 5G networks can be affected by several factors such as interference, congestion, signal attenuation, or attacks, which can lead to packet loss and retransmissions. Retransmissions in the network may be seen as an essential measure to improve network reliability, but a high retransmission rate may indicate issues that can help network operators mitigate possible service disruptions or threats to network users. A deep learning-based approach has been proposed to predict downlink retransmissions in 5G networks, achieving as much as 5%- 15% improvement over traditional prediction algorithms.