Modeling and analysis of spectrum handoff methods in cognitive radio networks

In cognitive radio networks, a licensed (primary) user must not be exposed of any interference by unlicensed users. Not damaging licensed users is a situation that cognitive radio must certainly provide. While an unlicensed user makes transmission in a frequency channel, the channel must be emptied if it is going to be used by a licensed user. The unlicensed user is required either to stop transmission or to continue transmission on another channel in order to free up the frequency channel. Channel changing of unlicensed users for emptying the frequency channel of licensed users is known as spectrum handoff. Various methods and techniques for performing spectrum handoff are available. By means of these methods and techniques, transmissions of unlicensed users can be transferred to another channel without interruption. In spectrum handoff process, a situation that should be taken into consideration is the priority classes. Priority classes determine the order of transmission among users. In other words, a user which has the highest priority has the right to transmit first. Another case that should be known about priority classes is utilizing either preemptive or non-preemptive priority techniques. In preemptive priority, high priority transmission is started by interrupting low priority transmission in the spectrum. In non-preemptive priority, a low priority transmission is waited to be finished, and high priority transmission is started after completion of low priority transmission. Priority classes in spectrum handoff process are chosen as thesis topic because it is an up to date and open workspace area. In this thesis, the design of the spectrum handoff process with priority classes for secondary users is provided, and in multi parameter decision making processes artificial intelligence techniques are utilized. With the help of designed channel bonding and priority based spectrum handoff method, the total number of handoff and the cumulative handoff delay are minimized while the total network throughput is increased.