I. Introduction
Spectrum is a requisite natural resource for wireless communication systems, which faces a serious shortage because of the explosive growth of data services nowadays. However, the phenomenon of low spectrum usage exists universally because of traditional fragmented spectrum allocation policy [1]. National Radio Network Research Testbed Project indicates the average utilization rate of spectrum below 3GHz is merely 5.2% [2]. In order to solve the contradiction between spectrum shortage and low spectrum utilization, cognitive radio (CR) technique is proposed, which can sense radio environment and utilize vacant spectrum flexibly [3].