An efficient filter for detecting gravitational wave bursts ininterferometric detectors

Typical sources of gravitational wave bursts are supernovae, for which no accurate models exist. This calls for search methods with high efficiency and robustness to be used in the data analysis of foreseen interferometric detectors. A set of such filters is designed to detect gravitational wave burst signals. We first present filters based on the linear fit of whitened data to short straight lines in a given time window and combine them in a non linear filter named ALF. We study the performances and efficiencies of these filters, with the help of a catalogue of simulated supernova signals. The ALF filter is the most performant and most efficient of all filters. Its performance reaches about 80% of the Optimal Filter performance designed for the same signals. Such a filter could be implemented as an online trigger (dedicated to detect bursts of unknown waveform) in interferometric detectors of gravitational waves