Cognitive radio networks for dynamic spectrum management.

摘要

A fundamental evolution is witnessed in recent research and development of wireless communications: much effort is put on shifting from the current static spectrum management to a dynamic one, since the wireless industry is hindered by spectrum scarcity while most of allocated spectrum stays idle. To realize the paradigm of dynamic spectrum management, cognitive radio network is a promising way. In this thesis, we present our current research status on dynamic spectrum management and cognitive radio networking, under a unified framework including both economic and technological aspects. We first investigate the economic feasibility of dynamic spectrum management in the form of dynamic spectrum market. Market mechanisms are proposed to stimulate primary (licensed) users to open up their under-utilized spectrum bands for secondary (unlicensed) users. We further analyze the competitive behavior of the major players in wireless industry, wireless service providers, in such a spectrum market. Then we switch to the technical realization part. We address several unique issues within cognitive radio networking. One is about the spectrum sensing strategy of secondary nodes. A MAC protocol is proposed to conduct efficient spectrum sensing and access in ad hoc cognitive networks, taking the hardware constraints into consideration. Another is about spectrum heterogeneity. We propose a cooperative relaying scheme and prove the effectiveness with a real testbed. Finally, we describe our current work about cognitive radio network testbed construction. The testbed has both fine programmability and high performance, which is quite useful for concept verification and performance evaluation.