Channel status prediction for cognitive radio networks

Abstract

The cognitive radio (CR) technology appears as an attractive solution to effectively allocate the radio spectrum among the licensed and unlicensed users. With the CR technology the unlicensed users take the responsibility of dynamically sensing and accessing any unused channels (frequency bands) in the spectrum allocated to the licensed users. As spectrum sensing consumes considerable energy, predictive methods for inferring the availability of spectrum holes can reduce energy consumption of the unlicensed users to only sense those channels which are predicted to be idle. Prediction-based channel sensing also helps to improve the spectrum utilization (SU) for the unlicensed users. In this paper, we demonstrate the advantages of channel status prediction to the spectrum sensing operation in terms of improving the SU and saving the sensing energy. We design the channel status predictor using two different adaptive schemes, i.e., a neural network based on multilayer perceptron (MLP) and the hidden Markov model (HMM). The advantage of the proposed channel status prediction schemes is that these schemes do not require a priori knowledge of the statistics of channel usage. Performance analysis of the two channel status prediction schemes is performed and the accuracy of the two prediction schemes is investigated.