Saturday, 20 February 2010

Structural properties of amorphous materials (PhD thesis)

Cofounder of Flying Frog Consultancy Ltd., Dr Jon Harrop, did his PhD at the University of Cambridge on the "Structural properties of Amorphous Materials" in the department of chemistry. A large part of this work was on the application of time-frequency analysis to the study of the diffraction data of amorphous materials. The complete PhD thesis is now freely available for download as a PDF here.

In the process, Dr Harrop developed a new approach to time-frequency analysis that is based upon the continuous wavelet transform (CWT). This method of analysis is applicable to a wide variety of signals that arise in many different areas of study including biology, physics, finance, engineering, and bioinformatics.

Chapter 2 introduces the method of time-frequency analysis and discusses various different approaches that have been invented over the years. In particular, the beneficial charactistics of the continuous wavelet transform are described in detail, along with the mathematics behind Delprat's time-frequency interpretation of the CWT.

Chapter 3 describes the evolution of a new approach to time-frequency analysis that avoids the problems of the existing methods. The new framework consists of an objective and quantitative measure of the suitability of a wavelet and the derivation of a wavelet that improves upon all others in the general case. The properties of the wavelet are then determined such that a more accurate time-frequency interpretation can be constructed, allowing the continuous wavelet transform with the new wavelet to be used as an accurate quantitative method of time-frequency analysis.

Creating a computational implementation that is fast, accurate and reliable requires considerable attention to detail. The software developed by Dr Harrop for the analysis of diffraction data has since been developed into a commercial Mathematica add-on package for general-purpose signal analysis.


Jonathan Shore said...

Interesting work. I think many have "hit a similar wall" with wavelet analysis, where one tends to lose signal information in parts of the signal towards the boundary of the wavelets.

I long ago moved away from wavelets to use other decomposition approaches due to similar problems. I have no studied your approach in detail to see how you address these.

Huang's HHT approach, for instance preserves the frequency and time domains without feature loss. I use another approach similar to his ...

emily said...

As you are aware of the latest progress of internet specially the introduction of web 2.0 websites is an online revolution. I appreciate your work to provide such authentic material.
buy thesis paper

osse said...

woow ! Very interesting post I like your website keep up the great posts
thesis papers to buy

jay paul said...

Nice post! Can’t wait for the next one. Keep stuff like this coming.

HP - Split 2-in-1 13.3" Touch-Screen Laptop - 4GB Memory - 128GB Solid State Drive - Modern Silver

HP - ENVY Touchsmart 14" Refurbished Touch-Screen Laptop - 8GB Memory - 500GB Hard Drive - Midnight Black