Adaptive learning in cognitive radio

Abstract

Machine learning is a powerful tool for cognitive radio users to learn its sensing and transmission strategy from the experience. This chapter provides a brief introduction to a variety of machine-learning techniques. The basic setup of machine learning, as well as the dichotomy, is explained. Then, the supervised, unsupervised, semi-supervised, and reinforcement learning techniques are briefly discussed. The single-agent learning is then extended to the case of multiagent learning. Then, the machine-learning techniques are applied in various cases of machine learning, such as channel selection and routing.