|
| Author: | Michael O'Brien |
| Institution: | 1. Science Applications International Corporation, 2. University of Hawaii |
| Email: | mob@gso.saic.com |
| Authors: | Michael O’Brien(1), Manoch Bahavar(1), Eli Baker(1), Milton Garces(2), Claus Hetzer(2), Hans Israelsson(1), Charles Katz(1), Colin Reasoner(1), and Jeff Stevens(1) |
Abstract Title:
"Detection of Infrasonic Events Using a Sparse Global Network of Infrasound Stations" |
Abstract:
We are working to improve and quantify the detection of infrasound events using data from the International Monitoring System (IMS) network of infrasound stations and similar stations in North America. We defined a project network as the set of IMS and US stations for which data were available to us as of July 1, 2003. Infrasound signal detection is complicated by persistently energetic and spatially coherent clutter, which can generate many false alarms that complicate event processing. We tuned the Detection and Feature Extraction (DFX)/libinfra program specifically for each station to achieve a manageable balance between the probability of detection (PD) of signals of interest and the false alarm rate (FAR). We are analyzing the characteristics of clutter for selected stations to assist in rejecting clutter detections, thereby decreasing the FAR without significantly compromising the PD. We are simulating the detection capability of the tuned project and IMS networks with the NetSim and Xnice programs, using inputs derived from the currently available IMS data: probability density functions of the ambient noise amplitude spectra estimated for data spanning January-August 2003; detection signal-to-noise ratio (SNR) thresholds estimated from Receiver Operator Characteristic curves that were computed by implanting signals in real station data; and properties for future IMS stations estimated from existing stations of comparable geography. Finally, we compiled a data set of events of interest to be used to assess detector performance. |
| Keywords: | IMS infrasound DFX FAR detection signal |
View the abstract list | Search the abstract list
|
