Date of Award

Summer 8-2016

Embargo Period

5-2-2019

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Electrical and Computer Engineering

Advisor(s)

Rohit Negi

Abstract

In order to support a multi-standard platform using a Software Defined Radio (SDR), the novel idea of reconfigurable Radio Frequency (RF) frontends have recently been proposed by the U.S. Defense Advanced Research Projects Agency (DARPA). A reconfigurable RF front-end has RF components that are reconfigured separately in order to satisfy the requirement of a particular communication standard. The reconfigurable RF front-end is a more reliable front-end for SDR than the currently used fixed wide-band RF front-ends, which have degraded system performance by passing more interference signals spread out in a wide range of frequency band. In order to realize the reconfigurable RF front-end, this thesis investigates the optimization method to select from the available configurations in radio environments with interference. In order to select an optimal configuration, we propose the Environment-Adaptable Fast (EAF) optimization algorithm for a reconfigurable RF front-end. A reconfigurable RF front-end not only needs to select an appropriate configuration that can operate for a given standard, but also needs to adapt quickly to a dynamic communication condition. This is difficult since there may be millions of available configurations. First, we studied RF impairment estimation for reconfigurable RF front-ends. Nonlinearity, phase noise, noise figure as well as frequency offset are RF impairments most likely to affect a particular standard. Second, we formulated the Signal-to-Interference-and-Noise Ratio (SINR) calculation which hastens the optimization process. We demonstrate the performance of the EAF optimization method in an exemplary scenario using Matlab Simulink. Finally, we designed the EAF optimization algorithm as a heuristic to select a configuration from the available ones. These simulation results demonstrate that while finding an optimal configuration, the EAF optimization significantly reduces simulation time compared to the four other previously proposed optimization methods. Thus, we expect that a reconfigurable RF front-end would be useful in real-time communication environments, since it would need significantly fewer reconfigurations to find an adequate configurations.

Available for download on Thursday, May 02, 2019

Share

COinS