Reinforcement Learning for Minimizing Communication Delay in Edge Computing

Abstract

For real-time edge computing applications working under stringent deadlines, communication delay between IoT devices and edge devices needs to be minimized. In order to minimize the communication delay between the IoT devices and the edge devices, we need a sophisticated approach for assignment IoT devices to the edge devices. Most of the heuristics solutions previously used to tackle the problem faced issues being solution stuck at local optima and high computational over head. To that end, researchers used reinforcement learning (RL) algorithms to explore the search space to get near optimal solutions. For our work, we consider RL based algorithms and show the preliminary results.