Secure Mobile Edge Server Placement Using Multi-Agent Reinforcement Learning

Mobile edge computing is capable of providing high data processing capabilities while ensuring low latency constraints power wireless networks, such as the industrial internet things. However, optimally placing servers (providing storage and computation services to user equipment) still a challenge. To place mobile in network, that network minimized load balancing performed on servers, we propose multi-agent reinforcement learning (RL) solution solve formulated server placement problem. The RL agents are designed learn dynamics environment adapt joint action policy resulting minimization servers. ensure adapted by maximized overall performance indicators, sharing information, experienced from each other network. Experiment results obtained analyze effectiveness proposed solution. Although information makes obtain network-wide maximation at same time it susceptible different kinds security attacks. further investigate issues arising solution, provide detailed analysis types attacks possible their countermeasures.