Growing demands of spectrum utilization make us to think about cognitive radio. Cognitive radio is smart enough to adjust its parameters according to spectrum use thereby increasing the efficiency of spectrum access. It means it should have the ability to make optimistic decisions dynamically in a multi operated network. Partially observable markov decision process is the technique of making optimistic decisions among various alternatives. MDPs are useful for studying a wide range of optimization problems solved via dynamic programming and reinforcement learning. In this paper, we demonstrate how POMDP is useful while using cognitive radio network.

Due to the rapid development of wireless communications in recent years, the demand on wireless spectrum has been growing dramatically, resulting in the spectrum scarcity problem. Works have shown that the fixed spectrum allocation policy commonly adopted today suffer from the low spectrum utilization problem. Both academic and regulatory bodies have focused on dynamic spectrum access to fully utilize the scarce spectrum resource. Cognitive radio, with the capability to flexibly adapt its parameters, has been proposed as the enabling technology for unlicensed secondary users to dynamically access the licensed spectrum owned by legacy primary users on a negotiated or an opportunistic basis. In this paper we present a volumetric survey on vari ous methods used to adapt changes used in cognitive radio.

No abstract