ar X iv : g r - qc / 0 40 41 23 v 1 2 9 A pr 2 00 4 Overview of the BlockNormal Event Trigger Generator

In the search for unmodeled gravitational wave bursts, there are a variety of methods that have been proposed to generate candidate events from time series data. Block Normal is a method of identifying candidate events by searching for places in the data stream where the characteristic statistics of the data change. These change-points divide the data into blocks in which the characteristics of the block are stationary. Blocks in which these characteristics are inconsistent with the long term characteristic statistics are marked as Event-Triggers which can then be investigated by a more computationally demanding multi-detector analysis. The ongoing search for the direct detection of gravitational waves using Earth based experiments involves the analysis of observations made at a variety of different detectors. These observations are time series samples of the detector state that are then processed by various means to identify gravitational wave candidates. Broadly speaking the searches for gravitational waves may be broken up into two categories, those searches that are based upon a model, such as the search for gravitational waves from inspiraling binaries, and those searches that endeavor to find gravitational waves without using a model. The latter category of searches are often referred to as " burst " searches. These searches typically seek to identify portions of data that are, for a short period, anomalously " loud " in comparison to the surrounding data. Identifying a gravitational wave burst in the absence of a source model is an involved and potentially computationally expensive process. This is especially true when the ratio of signal power to noise power is low. A convenient and natural approach to mitigating the computational expense of identifying such bursts divides the problem of detection into two parts. In the first part, an inexpensive procedure is used to identify candidate sections of data that trigger the second part of the analysis. The second part of the analysis focuses on the subintervals of data identified by the first. It is a more † current address: