Richard T. Lacoss

Microseisms: Mode Structure and Sources

Frequency-wave number spectra of microseisms were obtained by use of a set of short-period and long-period seismometers at LASA (Large Aperture Seismic Array, Montana). At times of relatively high microseismic activity short-period (shorter than 5 seconds) microseisms consist of both body waves and higher-mode surface waves. From the phase velocity and direction of body waves, source areas were determined, coinciding with low-pressure regions on the weather map. At longer periods, microseisms consist of fundamental- mode Rayleigh and Love waves, the former being dominant. Most microseismic energy arrives at LASA from the northeast and the west.

Use of persistent scatterers for model-based recognition

The signature of a target imaged by a millimeter-wave SAR is highly variable. Various viewing angles will cause different scattering centers to be illuminated, the returns from which can vary greatly with minor changes in viewing angle, and the coherence of the radiation induces speckle noise. Using fully polarimetric turntable (inverse SAR) data, we have undertaken some basic investigations of the persistence of scatterers as a function of azimuth for a number of depression angles from 15 degrees to 32 degrees. Although many scatterers persist for only a few degrees of azimuth, enough persist for 10 to 20 degrees to make model-based recognition feasible. Based on these results, we have developed an experimental system for target recognition. The system uses the functional template approach for detection, pose estimation, and initial hypothesis ranking. The best-matching template defines an area where so-called bright-points are extracted, resulting in a binary feature map that shows the location of strong scatterers. Back-end recognition consists of matching these feature maps to target appearance models that capture the location of scatterers that produce strong returns and are sufficiently persistent with changes in viewing angle. The performance of the hypothesis generation via functional templates is briefly reviewed, both for ISAR data and for SAR data. Recognition results obtained with the new back-end recognition system are also presented for the case of ISAR data.

Distributed Mixed Sensor Aircraft Tracking

A combination of geographically distributed small acoustic arrays and imaging sensors can be used to passively detect and track low-flying aircraft. Lincoln Laboratory has developed detection and tracking algorithms to illustrate this concept and has demonstrated them in real-time using an experimental test bed. This paper describes the algorithms, the test bed system and experimental results. The experiments are believed to represent the first real-time demonstration of a distributed mixed sensor tracking system of this type.

Distributed Estimation in the MIT/LL DSN Testbed

The acoustic tracking algorithms currently used in the MIT Lincoln Laboratory distributed sensor networks (DSN) testbed is described. The original motivation for inclusion of various features in those algorithms and the lesson learned about those features through experimentation with real and simulated data are discussed. Plans for modifications to the detection and tracking algorithms are sketched.

Processing partially coherent large seismic array for discrimination

Seismic signals in a band of frequencies near 1.0 Hz. can be detected from underground explosions and earthquakes at distances of several thousand kilometers. A stochastic model has been proposed to characterize observed signal variations within a Large Aperture Seismic Array (LASA). The model assumes that the observed signal spectrum at a seismometer is some average spectrum multiplied by a random gain and phase. Within a subarray (7 km. aperture) the mean value of the modulus squared of the random term is approximated by 1.0 +0.18 f2 where f is frequency in Hz. For sensors drawn from the full LASA (200 km. aperture) the value is 1.0 +2.0 f2. Two alternative methods for extracting spectral information above 1.0 Hz for discrimination between event types are compared. Beamforming spectra are obtained from the Fourier transform of the average received signal. An alternative incoherent processing method, spectraforming, is to calculate the average spectrum from individual seismometers. Both can be corrected for bias. It is demonstrated that although beamforming will give more noise rejection than spectraforming, that the latter can be superior in terms of output signal to noise ratio when input signal variations between sensors are large. Spectraforming may be of significant value for obtaining spectral information in the 1.0 - 3.0 Hz band for events with Richter magnitudes in the range 4.0 - 4.5. This magnitude range is of considerable current interest for the purpose of nuclear test detection and discrimination.