Question

I recently came across your "The GPS Tookit- Open Source Clock Tools." I am performing frequency metrology on microresonators for sensing applications, such as DNA based clinical diagnostics of disease. I am very interested in correctly characterizing the noise in these frequency measurements, which seems to entail utilizing a version of the Allan or Hadamard variances. Would you happen to have any Matlab code for calculating these variances that you would be willing to share? If I use it for a publication, I will definitely mention you in the acknowledgments or as a co-author, depending on how much the work depends on your code and if you are comfortable with being mentioned. Regardless, I would appreciate any guidance that you could offer me in correctly characterizing clock variance, as I am new to this field.

-- MattR - 19 Feb 2009

Answer

If you answer a question - or have a question you asked answered by someone - please remember to edit the page and set the status to answered. The status is in a drop-down list below the edit box.

The GPSTk Clock Tools computational code is written in C++. One can calculate Allan and overlapping Hadamard variance using Clock Tools. A Matlab implementation of the Dynamic Allan Variance is provided by Patrizia Tavella at http://www.inrim.it/res/tf/allan.shtml The paper CANVAS: Clock Analysis, Visualization, and Archiving System A New Software Package for the Efficient Management of Clock/Oscillator Data (http://ieeexplore.ieee.org/iel5/10510/33284/01574006.pdf?tp=&isnumber=&arnumber=1574006) by Ken Senior, Ronald Beard, and Joe White provides a link to Matlab implementations of the Allan and Hadamard Variances. Stable32 (http://www.wriley.com/) is a commercial package that allows for frequency stability analysis.

With respect to noise, are you wanting to identify power law noise types such as white pm, flicker pm, white fm, etc?

-- ColinPetersen - 20 Feb 2009