next up previous contents
Next: Multi-taper methods for spectral Up: GRASP Routines: General purpose Previous: Function: sound()   Contents


Example: translate

0 This example may be found in the src/examples/examples_utility directory of GRASP, and contains an example program which translates data from the ``old 1994" Caltech 40-meter format described earlier, to the new LIGO/VIRGO frame format. Because this code provides an example of how the data is encoded in this new format, we have included the text of the translation code here. The frames produced by this translation contain about 5 seconds of data each, and are about half a megabyte in length. The number of frames in each data file is set by the
# define FRAMES_PER_FILE
at the top of the code. To run the utility, use the command
translate directory-name
where directory-name is the name of the directory in which the files channel.0 to channel.15 may be found. The FRAME format files produced by translate are labelled uniquely by the time at which the first data point in the first frame was taken. The choice of file names depends upon the version of the Frame library. Here C1 denotes the Caltech 40-meter prototype. The names that will be used for the other sites are: H2 Hanford LIGO detectors (these will be stored in a single frame, together, with channel names to distinguish the 2km and 4km detectors, L for the Livingston LIGO detector V1 for the Virgo detector, G1 for the GEO detector, T for the Tama detector, S for the Glasgow detector, M for the Max-Plank detector, and A for the AIGO detector. In some cases (for example at Livingston and Hanford) additional frame files containing trend data, with typically one sample per channel per second, and one minute of data per frame, will be stored in files of the form H-609637094.T, with the suffix denoting ``trend". Includes/translate.tex
Author: Bruce Allen, ballen@dirac.phys.uwm.edu
Comments: The technique used to time-stamp this data is an attempt to correct the poor resolution of the original data - please see the remarks in [*] for additional detail. Also notice that because the sample rates of the slow/fast channels differ by a ratio of 10, we can not easily reformat the frames with sample sizes of length $2^n$. We expect that the FRAME format will continue to evolve, so that this translator (and the FRAME format data) may reqire periodic updates. Should the year have four digits (eg, 1994) for easier sorting?


next up previous contents
Next: Multi-taper methods for spectral Up: GRASP Routines: General purpose Previous: Function: sound()   Contents
Bruce Allen 2000-11-19