LIGO Data Analysis, Detector Characterization and Gravitational Wave Astronomy

This proposal addresses research in LIGO data analysis leading to gravitational wave astronomy. In particular, methods for removing instrumental noise from the LIGO data stream, preparatory to its analysis will be developed. This work, called 'data conditioning', is critical to the success of the LIGO since the gravitational wave signals are expected to be weak and would most likely be totally masked by instrumental noise. Several methods are proposed for conditioning LIGO data including the use of 'artificial intelligence' tools for time-ordered data. In addition, LIGO data will be analysed to

1. set upper limits on the gravitational wave strength from known pulsars,

2. search for unanticipated sources of gravitational waves,

3. combine data from the three LIGO detectors synthesized to perform as a single detector.

On technique for identifying gravitational wave signals in the data stream invloves the use of filters designed to pick out a signal with specific characteristics. Several innovative methods are proposed for the development of non-linear filters in the presence of non-Gaussian noise, involving the use of techniques to determine the most appropriate likelihood function for use when non-normal moments of the data distribution are known.