ABSTRACT

The future Cognitive Radio Networks (CRNs) will consist of heterogeneous devices such as smartphones, tablets and laptops moving dynamically. Accurate and robust spectrum sensing and identification of unauthorized spectrum usage are essential components of spectral efficiency in future radio systems. This project aims to utilize consensus-based cooperation featuring self-organizable and scalable network structure to capture the swarming behaviors of spectrum users and providing cooperative spectrum sensing in a fully distributed manner. By using a combination of control theory and machine learning techniques, the project designs secure weighted average consensus for cooperative spectrum sensing that can not only capture the swarming behaviors in CRNs with heterogeneous devices, but also is robust to practical channel conditions. Robust localization approaches are developed grounded on dynamic signal strength mapping, which have the capability to localize multiple malicious users. Additionally, the new techniques are validated using an actual testbed with on-campus deployment and system demonstration to industrial collaborators. The integration of control theory with dynamic spectrum access will enable a new revolution in the way for enhancing spectrum efficiency in CRNs. The project serves as a pioneer in exploiting multi-disciplinary knowledge (e.g., control systems and machine learning techniques) to achieve a more efficient spectrum usage in future radio systems, aiming to alleviate the increasing crowdness of the spectrum occupancy and support the co-existence of heterogeneous devices. This project also carries out a broad range of education and outreach activities to encourage students to pursue careers in the fields of science and engineering. Research results will be disseminated to academia and industry through presentations and publications in meetings, conferences and journals.

Please report errors in award information by writing to: awardsearch@nsf.gov.