NETWORK STUDIES-SIGNAL CHARACTERISTICS

• Main Authors: Benno, Stephen A, Johnson, William A, Strange, Peter L, Bonner, James A
• Other Authors: TEXAS INSTRUMENTS INC DALLAS SCIENCE SERVICE DIV
• Format: Text
• Language: English
• Published: 1968
• Subjects: Seismology; Acoustics; *SEISMOLOGICAL STATIONS; BACKGROUND; CORRELATION TECHNIQUES; DATA PROCESSING; DENSITY; DETECTION; EARTHQUAKES; ELECTROMAGNETIC WAVE FILTERS; NETWORKS; NOISE; NUCLEAR EXPLOSIONS; PHASED ARRAYS; POWER SPECTRA; SAMPLING; SEISMOMETERS; SIGNAL TO NOISE RATIO; SIGNALS; SPECTRUM SIGNATURES; BEAM STEERING; SEISMIC NOISE; VELA UNIFORM; Kamchatka
• Online Access: http://www.dtic.mil/docs/citations/AD0830463; http://oai.dtic.mil/oai/oai?&verb=getRecord&metadataPrefix=html&identifier=AD0830463

The report presents an initial study of signal characteristics and coherent signal processing at the network level. Topics investigated include signal similarity, depth-phase detection and recognition, separation and location of events overlapping in time and space, and methods for real-time network processing and data presentation. Large variations in signal waveform and signal-to-noise ratio (SNR) are observed across the network. Interstation coefficients of correlation for a relatively simple Kamchatka event vary from 0. 54 to 0.96, with the higher correlation found for stations on the eastern North American continent. Levinson equalization filtering does not appear particularly effective at the network level in terms of improved correlation coefficients or SNR. Depth-phase detection at the network level is not materially better than for selected stations, but recognition appears more reliable than for general station results. Experiments show that overlapping events of differing magnitudes and separated in epicenter by as little as 1 degree can be resolved by network beamsteer and integrate techniques. Signal-enhancement procedures implemented primarily to determine the extent of signal attenuation caused by signal dissimilarity across the network indicate less than 3-db reduction in signal energy across the signal passband.