Autoregressive Spectrum Hole Prediction Model for Cognitive Radio Systems

Abstract

In this paper, an autoregressive channel prediction model is presented for cognitive radio(CR) systems to estimate spectrum holes. This model adopts a second-order autoregressive process and a Kalman filter. A Bayes risk criterion for spectrum hole detection is presented by considering interference temperature and channel idle probability. Theoretical analysis and simulations show that CR systems based on this scheme can greatly reduce the number of collisions between licensed users and rental users.