SpecEES: Collaborative Research: Energy Efficient Dynamic Spectrum Access in Uncoordinated Networks

University Of Illinois At Urbana-Champaign, Urbana IL

Investigators

Abstract

With its focus on energy efficient dynamic spectrum access where the users are treated as equals and there is minimal coordination across the users, this project addresses challenges for enabling next-generation wireless communications and networking in a dynamic spectrum environment. The results from the project should greatly enhance access to broadband wireless services in traditionally under-serviced communities. The spectrum-sensing and spectrum-sharing techniques developed should prove to be equally applicable in a slew of applications involving energy-constrained ad hoc networks such as those encountered in disaster relief operations and in a variety of applications relevant to national security. The involvement of undergraduate students in this project, especially in software radio experiment, is expected to foster their long-term interest in Science, Technology, Engineering and Math (STEM)-related fields. This research project is focused on both spectral efficiency and energy efficiency for a class of dynamic spectrum access applications that have been much less well studied: an uncoordinated wide-band system where the users are treated as equals, i.e., there is no primary/secondary distinction. The project will develop the spectrum management framework as well as enabling technologies for such applications, ranging from multi-channel spectrum sensing using quickest change detection to novel spectrum access schemes for spectral/energy efficient and fair spectrum usage in the absence of user coordination. Experimental studies are an integral part of the proposed research, and the project leverages existing software radio capabilities developed at Syracuse University. This study of spectrum access with equally privileged users without coordination also fills an important gap in dynamic spectrum access. Given the generality of the models considered, this research can be expected to make fundamental contributions to the theories of quickest change detection and random access.