AFIT neural network research

A brief summary of research done at the Air Force Institute of Technology (AFIT) in the area of neural networks is provided. It has been shown that backpropagation, used for feedforward artificial neural networks, is just a degenerate version of an extended Kalman filter, and that networks can do about as well as the optimum statistical classification technique. A method of finding the importance of features for use by a neural network classifier has been determined. Techniques for using neural networks for image segmentation have been developed. In optical pattern recognition, techniques that allow the processing of real FLIR (forward-looking infrared) images with existing binary spatial light modulators have been devised. An optical direction of arrival detector applicable to laser illumination direction determination has been designed and tested; the design is similar to a fly's eye. Coated mirrors for the optical confocal Fabry-Perot interferometer have been designed, specified, fabricated, and installed. Significant progress has been made in the use of neural networks for processing multiple-feature sets for speech recognition     

Practical stabilization in attitude thruster control

The attitude stabilization of a spacecraft using thrusters is considered from a practical point of view, i.e., when actuator constraints, uncertainties and failures, measurement noise, fuel consumption, and inertia matrix uncertainty are considered. A variable structure controller based on sliding modes is obtained first, which guarantees global exponential stability under actuator constraints, uncertainties, and failures. The design is based on a continuous average-proportional torque selection function and a set of feasible sliding surfaces, which may be adapted to improve performance. Exponential convergence towards ultimate bounds is guaranteed when all other practical issues arise.     

A wavelet multiresolution analysis for spatio-temporal signals

The wavelet filters of the conventional 3D multiresolution analysis possess homogeneous spatial and temporal frequency characteristics which limits one's ability to match filter frequency characteristics to signal frequency behavior. Also, the conventional 3D multiresolution analysis employs an oct-tree decomposition structure which restricts the analysis of signal details to identical resolutions in space and time. This paper presents a 3D wavelet multiresolution analysis constructed from nonhomogeneous spatial and temporal filters, and an orthogonal sub-band coding scheme that decouples the spatial and temporal decomposition processes     

Simulation of fast ablation

A brief review of studies on simulation and control of a porous cooling process taking into account material transpiration is given. An approximated solution for an equation of the one-dimensional problem of fast ablation is also presented.     

How much of the solar irradiance variations is caused by the magnetic field at the solar surface?

The contribution to total solar irradiance variations by the magnetic field at the solar surface is estimated. Detailed models of the irradiance changes on the basis of magnetograms show that magnetic features at the solar surface account for over 90% of the irradiance variations on a solar rotation time scale and at least 70% on a solar cycle time scale. If the correction to the VIRGO record proposed by Fröhlich & Finsterle (2001) is accepted, then magnetic features at the solar surface are responsible for over 90% of the solar cycle irradiance variations as well.     

Design and implementation of near-field, wideband synthetic aperture beamformers

A coarray-based near-field, wideband synthetic aperture beamformer using stepped-frequency signal synthesis and post-data acquisition processing is presented. While coarray techniques offer significant reduction in the number of array elements for a given angular resolution, the hybrid subarray-stepped frequency realization of wideband systems simplifies implementations and offers flexibility in beamforming. Proof of concept is provided using real data collected in an anechoic chamber for several pulse shapes and array weightings.     

Fiber optic fence sensor developments

Many detection technologies have been employed for perimeter detection, to sensitize a barrier. These outdoor perimeter fence detection sensors must reliably detect intruders attempting to cut or climb the barrier, while ignoring the effects of environmental noise, including nearby activity. In recent conference proceedings, the new IntelliFIBER/spl trade/ fiber optic based product was introduced and compared with previous technologies. IntelliFIBER is designed to provide the advantages of a nonconductive "dielectric" cable sensor, e.g., resistance to electromagnetic interference and the ability to provide longer cable zones. It utilizes the proven processor of the Intelli-FLEX/spl trade/ triboelectric cable fence sensor, which provides relay or bidirectional networked serial communications and is compatible with its calibration module. IntelliFIBER also capitalizes on the Intelli-FLEX's adaptive detection algorithms that were developed through extensive field-testing, to provide exceptional immunity to environmental alarms while still detecting the skilled intruder.     

Real-time translational control of a MEMS comb resonator

Closed-loop control has been successfully applied to a microelectromechanical systems (MEMS) lateral comb resonator device in real-time to perform impulse disturbance damping and sinusoidal position control, enabled by the use of a through-wafer optical microprobe to obtain position feedback. This result leverages the application of lifetime, in-situ control of MEMS in order to provide quality assurance of microsystems in safety critical applications. A position feedback signal produced by a through-wafer optical microprobe has been used for comb resonator system model identification by two independent methods to accurately determine the effective mass, damping, and spring constant values of the device. After accurate determination of system parameters, closed-loop impulse disturbance damping and proportional-integral-differential (PID) translational control were applied. Closed-loop control results presented indicate controllability of such microstructures and response times on the order of the natural frequency of the device.     

Parametric Rao Test for Multichannel Adaptive Signal Detection

The parametric Rao test for a multichannel adaptive signal detection problem is derived by modeling the disturbance signal as a multichannel autoregressive (AR) process. Interestingly, the parametric Rao test takes a form identical to that of the recently introduced parametric adaptive matched filter (PAMF) detector for space-time adaptive processing (STAP) in airborne surveillance radar systems and other similar applications. The equivalence offers new insights into the performance and implementation of the PAMF detector. Specifically, the Rao/PAMF detector is asymptotically (for large samples) a parametric generalized likelihood ratio test (GLRT), due to an asymptotic equivalence between the Rao test and the GLRT. The asymptotic distribution of the Rao test statistic is obtained in closed form, which follows an exponential distribution under the null hypothesis H ₀ and, respectively, a noncentral Chi-squared distribution with two degrees of freedom under the alternative hypothesis H ₁. The noncentrality parameter of the noncentral Chi-squared distribution is determined by the output signal-to-interference-plus-noise ratio (SINR) of a temporal whitening filter. Since the asymptotic distribution under H ₀ is independent of the unknown parameters, the Rao/PAMF asymptotically achieves constant false alarm rate (CFAR). Numerical results show that these results are accurate in predicting the performance of the parametric Rao/PAMF detector even with moderate data support.     

TerraSAR-X active radar ground calibrator system

In April 2006, the TerraSAR-X satellite was launched. This paper describes the development of a novel and highly integrated, digitally-controlled active SAR system calibrator (DARC). It consists of both an active transponder path for absolute radiometric calibration and a calibrated receiver chain for antenna pattern evaluation of the satellite antenna. A total of 16 active transponder and receiver systems and 17 receiver-only systems will be fabricated for a calibration campaign in 2006.