Deep Learning-Based Protocol Stack Optimization in High-Density 5G Environments
Keywords:
Wireless Technology, 5G Networks, Internet of Things, Smart Cities, Millimeter Wave, High-Density Networks, Multi-Tier Deployment, Low Latency, Energy Efficiency, Protocol Stack Optimization, TCP/IP Layer, Deep Learning, Reinforcement Learning, Congestion Control, Network Slicing, Massive Machine-Type Communication, Mobile Broadband, Spectrum Protection, AI in Networking, Dense IoT Environments.Abstract
In this context, artificial intelligence and, more specifically, machine learning algorithms are being deployed to help automatize, optimize, and innovate the methods being applied. Focusing on the network protocol stack, we consider a Deep Learning approach to optimize the core TCP/IP layer for 5G environments when the terrestrial network is supporting large numbers of connected devices, running massive machine-type communications and multimedia-enhanced mobile broadband applications utilizing network slicing. In particular, considering a specific scenario where deep reinforcement learning is employed to control and optimize TCP protocol congestion window size, we show that Deep Learning-Based approaches might facilitate TCP optimization and avoid congestion in 5G networks, leading the way towards the implementation of smart solutions for protocol stack overhead use in dense IoT wireless environments.
