Dynamic channel selection with reinforcement learning for cognitive WLAN over fiber
International Journal of Communication Systems
| Hi-index | 0.00 |
The performance of legacy WLAN systems will significantly degrade as the number of active stations (STA) in the network increases, rendering satisfactory coverage of busy hot spots challenging in practice. A novel centralized architecture that coordinates several WLAN access points is the exploitation of Radio over Fiber (RoF) technology. In this paper we focus on spatial diversity techniques to enhance the performance of the newly proposed Cognitive WLAN over Fiber (CWLANoF) compared to legacy WLAN Extended Service Set (ESS). More specifically we develop an optimal transmitter-side beamforming technique and investigate the TCP performance enhancement as a result of usage of this scheme. Furthermore, the issue of fairness will also be discussed, where we introduce a fairness measure based on the distribution of TCP throughput among existing STAs in the network. The numerical results verify the superior performance of CWLANoF compared to WLAN ESS.