There has been an exponential increase in spectrum demands due to new emerging wireless services and applications, making it harder to find unallocated spectrum bands for future usage. This potential resource scarcity is rooted at inefficient utilization of spectrum under static spectrum allocation. Therefore, a new concept of dynamic spectrum access (DSA) has been proposed to opportunistically utilize the legacy spectrum bands by cognitive radio (CR) users. Cognitive radio is a key technology for alleviating this inefficient spectrum utilization, since it can help discover spectrum opportunities (or whitespaces) in which legacy spectrum users do not temporarily use their assigned spectrum bands.

In a DSA network, it is crucial to efficiently identify and utilize the whitespaces. We address this issue by considering spectrum sensing and resource allocation. Spectrum sensing is to discover spectrum opportunities and to protect the legacy users (or incumbents) against harmful interference from the CR users. In particular, sensing is an interaction between PHY and MAC layers where in the PHY-layer signal detection is performed, and in the MAC-layer spectrum sensing is scheduled and spectrum sensors are coordinated for collaborative sensing. Specifically, we propose an efficient MAC-layer sensing scheduling algorithm that discovers spectrum opportunities as much as possible for better quality-of-service (QoS), and as fast as possible for seamless service provisioning. In addition, we propose an optimal in-band spectrum sensing algorithm to protect incumbents by achieving the detectability requirements set by regulators (e.g., FCC) while incurring minimal sensing overhead.

For better utilization of discovered spectrum opportunities, we pay our attention to resource allocation in the secondary spectrum market where legacy license holders temporarily lease their own spectrum to secondary wireless service providers (WSPs) for opportunistic spectrum access by CR users. In this setting, we investigate how a secondary WSP can maximize its profit by optimally controlling the admission and eviction of its customers (i.e., CR users). In addition, we also focus on the price and quality competition between co-located WSPs where they contend for enticing customers by providing more competitive service fee while leasing the channels with best matching quality.