Forward Backward autoregressive spectrum prediction scheme in Cognitive Radio Systems

This paper proposes a simplified spectrum prediction scheme based on AR modeling, namely, Forward Backward AR (FBAR) model, that contributes to the field of Cognitive Radio (CR) through enhancing the prediction accuracy and the Secondary user throughput as compared to the existing techniques. Simulation Results show that the proposed method outperformed the nominal performance measures in literature with much reduced computational complexity.