Dynamic Spectrum Handoff for Cognitive Radio Sensor Networks Using Hidden Markov Models

To address efficient spectrum utilization challenge, cognitive radio has emerged as the key technology, which enables opportunistic access to the spectrum. In this paper, we propose a novel proactive spectrum handoff scheme for cognitive radio sensor networks, using Hidden Markov Model (HMM) to model and predict the spectrum usage of licensed bands. Based on this model, we estimate the channel idle time in the near future which plays a key role in proposed proactive spectrum handoff method. Experimental results show that the proposed approach for spectrum prediction is more effective compared with existing counterparts, adopting random selection, classification-based selection and probability-based selection respectively.