Authors:
Mariam Musavi, Kok LIm Alvin, Hafizal Mohamad & Nordin Ramli
Abstract:
Cognitive radio (CR) is a promising next-generationwireless communication system that provides efficient utilizationof radio spectrum by enabling unlicensed users (or secondaryusers, SUs) to sense for and use underutilized radio spectrum(or white spaces) owned by licensed users (or primary users,PUs). In this paper, we investigate the effects of a larger networksize (or higher number of routes) on non-clustered, clusteredand clustered-reinforcement learning (RL)-based route selectionschemes in a USRP/ GNU radio platform focusing on thenetwork layer. Experimental results show that the enhancedvariant of reinforcement learning (RL)-based route selectionscheme (C-ERL) selects stable route(s) over a clustered CRN in aUSRP/ GNU radio platform. C-ERL improves cluster stability byreducing the number of route breakages caused by route switches,and network scalability by reducing the number of clusters inthe network without significant deterioration of QoS, includingthroughput, packet delivery rate, and end-to-end delay
Source:
16th IEEE Asia Pacific Wireless Communication Symposium (APWCS 2019), Singapore
