Handoff reduction based on prediction approach in Cognitive Radio Networks

Abstract

Cognitive Radio Network is a key solution to improve the spectrum utilization by enabling unlicensed users (Secondary Users) to exploit the spectrum in an opportunistic manner. Since secondary users are temporary visitors to the licensed spectrum (Primary Users), they should vacate the spectrum once a primary user reclaims it. This process has a large overhead for secondary users. This overhead includes spectrum sensing and channel decision as well as communication delay between secondary users. To decrease this overhead it is necessary to reduce the number of handoff as much as possible. Increasing handoff may lead to low throughput of secondary users. In this paper, a dynamic handoff procedure for cognitive radio network is proposed based on decreasing handoff rate as well as increasing overall secondary users' throughput. With the aim Hidden Markov Model (HMM), each secondary user predicts the next primary user's arrival based on past observation and a model which is learned by past data. Based on this model, SUs can switch their operating channel to another before occurring collisions with primary user. Also, a SU may decide to stay at the current channel to decrease the overall latency. Simulation result shows that the proposed approach can meet QOS requirements.