next up previous index
Next: P7.12 Using Python in Up: Session P7. Software Systems Previous: P7.10 Automated Programming and   Author Index


P7.11 Automated Programming and CMB Signal Compression Simulations for Planck/LFI

Michele Maris ( Trieste Astronomical Observatory (OAT) - Trieste - Italy )

Fabio Pasian ( OAT )

Riccardo Smareglia ( OAT )

Davide Maino ( SISSA - Via Beirut 1/A, I-34131 - Trieste - Italy )

Maria Staniszkis ( Mathematics and Informatics Department - Udine University - Italy )

Carlo Burigana (TESRE - Bologna - Italy )

Jose Barriga (IEEC - Barcelona - Spain )

Enrique Gaztañaga (IEEC - Barcelona - Spain )

Reference URL: http://pv.infn.it/~maris/apgatpl.html

The concept of Automatic Programming and Managing of simulations is applied to an experimental problem: simulations of Cosmic Microwave Background (CMB) data compression in the framework of the ESA Planck/LFI mission. The LFI instrument is planned to fly on the Planck mission in 2007 and will produce full sky CMB maps over four frequency channels in the 30 - 100 GHz band. The limited bandwidth reserved to the downlink of scientific data calls for huge compression factors. In addition, the compression has to be lossless becouse the CMB signal is statistically similar to gaussian noise. The theoretical upper limit for the compression factor in these conditions is a bit less eight, but this limit has to be fixed more carefully with ``realistic'' sinthetic data and true compressors. Various compression strategies may be applied, and the selection a strategy of choice has to be firmly grounded on an exaustive set of tests with simulated data. To accomplish these tasks a great number of simulations of the CMB signal compression were carried out taking in account the full set of mission parameters, physical parameters and compression methods. In this context an Automated Program Generator was applied to automatize most of this ``explorative'' simulation process which includes: CMB signal generation, detection simulation, compression, evaluation and visualization of data and results. Automated simulation methods save time, reduce errors and simplify the simulations management and documentation. This communication illustrates the software architecture developed for this case and the main results.


next up previous index
Next: P7.12 Using Python in Up: Session P7. Software Systems Previous: P7.10 Automated Programming and   Author Index
adass@ncsa.uiuc.edu