Trading Spectrum Ownership for Performance

In the last decades, the unlicensed spectrum bands have become crowded, due to a proliferation of innovations in wireless communications and networking. In contrast, the licensed spectrum is considerably under-utilized. These two observations have motivated the industry and the spectrum authority to promote dynamic spectrum access so that unlicensed users can dynamically detect and access idle licensed spectrum bands, to significantly increase spectrum utilization and encourage innovations in wireless communications.

However, the existing dynamic spectrum access paradigm imposes a `foe' relationship between primary users and secondary users (secondary users), where the spectrum is exclusively used by either primary users or secondary users. In particular, a secondary user may access the spectrum only when primary users cease using the spectrum, and the resurgence of primary user traffic disrupts secondary user traffic. This paradigm does not offer sufficient incentives to primary users for embracing dynamic spectrum access. Furthermore, to comply with stringent dynamic spectrum access policies, secondary users are demanded to behave conservatively to sense and access the spectrum, resulting in an actual spectrum utilization significantly lower than what is feasible. In this project, we propose new architectures for dynamic spectrum access to offer incentives to primary users, so that both primary users and secondary users significantly benefit from dynamic spectrum access. We further design algorithms and protocols to improve the performance of dynamic spectrum access in cognitive radio networks.