Introduction to Reinforcement Learning
Introduction to Reinforcement Learning
Semi-dynamic simulator for large-scale heterogeneous wireless networks
International Journal of Mobile Network Design and Innovation
Fuzzy inference system learning by reinforcement methods
IEEE Transactions on Systems, Man, and Cybernetics, Part C: Applications and Reviews
A fuzzy reinforcement learning approach to power control in wireless transmitters
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Autonomic downlink inter-cell interference coordination in LTE self-organizing networks
Proceedings of the 7th International Conference on Network and Services Management
A fuzzy reinforcement learning approach for pre-congestion notification based admission control
AIMS'12 Proceedings of the 6th IFIP WG 6.6 international autonomous infrastructure, management, and security conference on Dependable Networks and Services
Hi-index | 0.00 |
Self-organizing networks (SONs) are considered as a driving technology that aims at enhancing usage of radio resources, at simplifying network management, and at reducing cost of operation of next generation radio access networks. This paper describes a framework for designing SON mechanisms for dynamically optimizing Radio Resource Management (RRM) functions. The base station is modeled as an agent that learns from its own local information and that of its neighbors to dynamically optimize RRM parameters. An application of the design framework to SON enabled fractional power control (FPC) in a LTE network is presented. The FPC is particularly important in OFDMA technology as a means to mitigate interference originated by uplink transmission power between neighboring cells. The agent uses fuzzy-reinforcement learning to dynamically adjust the FPC parameter to reach optimal tradeoffs between cell-edge and neighboring cell performance. The learning process is adapted to operate in a sporadic context related to the rapid variations in power, in users' position and in the number of interferers. Results show important gain brought about by the self-optimizing FPC to the network capacity and to the perceived quality for data applications.