Abstract
In cognitive wireless networks, spectrum owners (primary users, PUs) may lease the unused spectrum to unlicensed users (secondary users, SUs). This spectrum is used to establish a secondary network that serves real time connections. The size of leased spectrum influences both the admitted traffic of SUs and the cost of spectrum. For this spectrum market, we present unsupervised learning paradigm as a means for extracting the optimal control policy for spectrum trading. This policy gives spectrum owner the opportunity to maximize its profit by adapting network resources to the changes in the network status and the market conditions. To meet different requirements, the problem is formulated as reward maximization with penalty for delay. The numerical results show that the proposed machine learning method is able to find an efficient trade-off between profit loss, and average delay for SUs.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call