Abstract： While more and more devices go wireless, the demand for more radio spectrum bandwidth increases dramatically. In contrary to the general opinions that we will run out of the radio spectrum resources quickly, it is shown that the usage of most licensed bands is low and a large portion of the spectrum resources is wasted. The cognitive radio technology allows the unlicensed users to co-exist seamlessly with the licensed users in order to efficiently use the precious radio spectrum resources. A cognitive radio device first observes the radio environments by performing spectrum sensing. Based on the sensing results, the cognitive radio decides when and how to access the channels by choosing the most appropriate transmission scheme. In this talk, we will present our proposal of the functional architecture of a cognitive radio system, investigate its components, and discuss our approaches. Due to the interdisciplinary nature of the cognitive radio, the system requires the integration from various fields of communications, signal processing, machine learning, and hardware management. As a result, this talk hopes to stimulate the applications of new methods from other fields to the wireless communications and the cognitive radio research. Bio Chia-han Lee received his B.S. degree from National Taiwan University, Taipei, Taiwan in 1999, M.S. degree from the University of Michigan, Ann Arbor in 2003, and Ph.D. from Princeton University in 2008, all in electrical engineering. From 1999 to 2001, he served in the R.O.C. army as a missile operations officer. Since 2008, he has been a postdoctoral research associate at the University of Notre Dame. His research interests include wireless communications, cognitive radio, machine learning, error correcting codes, and stochastic geometry.