Spectrum occupancy prediction using a hidden Markov model

Spectrum occupancy prediction is a key enabler of agile, and proactive spectrum utilization in dynamic spectrum access networks. Bayesian-based techniques manifested by Hidden Markov Model provide powerful, and flexible tools for statistical spectrum prediction. In this paper, we simulate the performance of single step-ahead prediction, in terms of observation process errors, and state transition probability. We model the primary, and the secondary users shared spectrum channel as a two state hidden Markov model. Mean prediction error is calculated, and presented as a function of the model parameters.