Skip to main content
eScholarship
Open Access Publications from the University of California

UC San Diego

UC San Diego Electronic Theses and Dissertations bannerUC San Diego

Information processing with longitudinal spectral decomposition of ultrashort pulses

Abstract

Optical signal processing with ultrashort pulses allows for the synthesis and analysis of signals at speeds exceeding the limits of conventional electronics. The origin of this processing capability is the large, well- phased bandwidth required to support short optical pulse durations. In comparison with ultrafast optical signals, wideband radio-, micro-, and millimeter wave signals are relatively narrowband and therefore are readily created or detected through optical techniques. Researchers traditionally utilize Fourier synthesis methods to operate on broadband optical signals. Such an approach relies on decomposing the optical spectrum and manipulating it with a mask or filter. While the spatial domain is most commonly employed for such processing, this approach suffers from slow update speeds and must scale in volume to increase signal complexity. This dissertation explores an alternative approach relying on the decomposition of the optical spectrum of ultrashort pulses in the time domain using chromatically dispersive fiber technologies. The approach is coined longitudinal spectral decomposition in order to contrast with the traditional transverse spectral decomposition. The dissertation is organized to first familiarize the reader with the toolbox of technologies and signal processing techniques available for the creation and modification of longitudinal spectral decomposition waves. The primary distortion to such processing, associated with higher order dispersion, is introduced and theoretically treated up front. Subsequently, a sequence of applications driven by longitudinal spectral decomposition is presented. These applications include: optical pulse shaping, microwave spectrum analysis and signal generation, as well as high speed optical reflectometry or ranging. By presenting these applications, the dissertation highlights the processing advantages of longitudinal spectral decomposition while also covering a number of subtle issues associated with the method and its supporting technologies. The goal of the work is to illustrate to the reader the unique capabilities enabled by processing ultrashort pulses in the time domain. Future improvements to broadband optical source generation, dispersive fiber element fabrication, and optical/electrical signal interfacing promise to increase the efficacy of applications relying on longitudinal spectral decomposition and to extend the viability of the technique as a stand-alone processing platform

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View