Autonomous Resource Allocation and Strategy Optimization in Wireless Networks
In this paper, we propose an autonomous control framework, where the focus remains on autonomous resource utilization and strategy optimization. The control framework aims at an immensely dynamic, time varying, and less predictable network environment. The first main motivation of the proposal is to eliminate redundant resource utilization for specific network objectives while reducing the network response time. To this end, the decision making engines of the control framework characterize the network environment in terms of the correlation with specific network objectives and improve the resource utilization accordingly. The second and the last motivation is to match network strategies with different environmental conditions in terms of impact rates on the environment. We make use of Partially Observable Markov Decision Processes (POMDP) based control loop to improve the network strategies against alarm conditions for different environmental conditions. The control framework is implemented on WLAN APs and the proposed approach is validated by means of real implementation and simulation of different network scenarios.