PBRAMEC: Prioritized Buffer Based Resource Allocation for Mobile Edge Computing Devices

In recent years, cloud-based smartphone applications like augmented reality (AR), facial recognition, and object detection have gained popularity because the remote execution of cloud computing may create significant latency and increase back-haul bandwidth usage. Addressing these issues the research seeks to employ Deep Deterministic Policy Gradient (DDPG), type of Reinforcement Learning (RL) and enhance it by prioritizing the experiences stored int the replay buffer to allocate resources for mobile users in an edge computing environment. Edge computing, which proceeds storage and processing resources near the mobile users, can increase reaction times and relieve back-haul congestion by taking into account the computational resources, migration bandwidth, and offloading targets.