This project studies non-intrusive opportunistic access in spectrum-agile communication networks. This work is motivated by the need for efficient spectrum utilization, facilitated by regulatory policy movements, and enabled by advances in hardware technologies. The investigators focus on sensing-based secondary networks to achieve non-intrusiveness because sensing-based schemes require low infrastructure support and can complement other approaches. The following issues are studied: 1) capacity-interference tradeoff between primary and secondary users; 2) evacuation of secondary users upon the return of primary users; and 3) compatibility with CSMA-based primary users. The project involves theoretical analysis, system modeling, protocol design, and experimental evaluation. An integrated approach is taken that involves physical layer, MAC layer, and network layer and spans from theory to practice. The investigators exploit tools from estimation and detection, graph, and optimization theories for analysis and modeling, and design experiments for performance evaluation and model validation. The results of the project include feasibility and capacity analysis of sensing-based approach, including both non-interactive (e.g., TDM/CDM-based) and interactive (e.g., CSMA-based) primary systems; and protocol suites that evacuate secondary users fast and reliably based on the interference tolerance limit of primary users. The outcome of this project will provide the research community with good understandings on non-intrusive spectrum-agile communication systems and policy makers with theoretical limits and experimental data. Protocols and models developed and validated in this project can be used by other researchers as building blocks of spectrum-agile systems. The project also enhances the education curriculum and fosters the interdisciplinary collaborations.
