Downloads & Free Reading Options - Results

A number of results were obtained during the period 1 August 1979 to 30 July 1980. Studies were made of the implementation of narrowband digital filters with reduced roundoff error and limit cycle oscillations. Bounds on limit cycles of second order filters were derived for filters using error spectrum shaping. The performance of two sequential signal detectors were analyzed. A schemee for the computer generation of random sequences with a specified first order distribution and a specified autocovariance function is proposed and analyzed. A delta modulation like sampled analog filter structure is developed. The filter uses only coefficients O,+1, and -1, and can be fabricated as a programmable CCD filter. Work has begun on a further refinement of narrowhead spectrum analysis by direct decimation and on the use of recursive filters for decimation and interpolation.

A number of results were obtained during the period 1 August 1979 to 30 July 1980. Studies were made of the implementation of narrowband digital filters with reduced roundoff error and limit cycle oscillations. Bounds on limit cycles of second order filters were derived for filters using error spectrum shaping. The performance of two sequential signal detectors were analyzed. A schemee for the computer generation of random sequences with a specified first order distribution and a specified autocovariance function is proposed and analyzed. A delta modulation like sampled analog filter structure is developed. The filter uses only coefficients O,+1, and -1, and can be fabricated as a programmable CCD filter. Work has begun on a further refinement of narrowhead spectrum analysis by direct decimation and on the use of recursive filters for decimation and interpolation.

Tracking performance depends upon the quality of the measurement data. In the Kalman-Bucy filter and other trackers, this dependence is well-understood in terms of the measurement noise covariance matrix, which specifies the uncertainty in the values of the measurement inputs. The measurement noise and process noise covariances determine via the Riccati equation, the state estimation error covariance. When the origin of the measurements is also uncertain, one has the widely-studied problem of data association (or data correlation), and tracking performance depends critically on additional parameters, primarily the probabilities of detection and false alarm. In this paper we derive a modified Riccati equation that quantifies (approximately) the dependence of the state error covariance on these parameters. We also show how to use a ROC curve in conjunction with the above relationship to determine an optimal detection threshold in the signal processing system that provides measurements to the tracker. A validation of the modified Riccati equation is also presented.

In this effort, we considered the problem of classifying closely spaced UXO. When two UXO are in close proximity, their signatures as measured by electromagnetic induction (EMI) sensors co-mingle, and traditional classification algorithms cannot be utilized. We explored the use of independent components analysis (ICA), a technique from the blind source separation (BSS) literature as a pre-processing step by which to separate the individual UXO signatures from the mixtures measured by the EMI sensor. Simulations indicated that this procedure could succeed and restore some level of classification performance in the case of overlapping signatures. Test-stand data also indicated that the ICA-based classification approach showed promise. Finally, testing the algorithms on data collected by NRL/WES suggested that in some cases ICA could be used as a preprocessing step and that closely-spaced UXO could be classified. During the course of this effort, several studies were performed to try to assess the limitations of the ICA-based approaches for discrimination of closely spaced objects. Correlation between object signatures can degrade performance, and alternate techniques based on ICA were considered as a remedy to this issue. Other issues including sampling density, noise susceptibility, and others were also investigated.

The long-term goal of this project is to develop fast and accurate methods for geoacoustic inversion, localization, and detection by combining the physics of sound propagation in the ocean and principles of signal processing.

The long-term goal of this project is to develop fast and accurate methods for geoacoustic inversion, localization, and detection by combining the physics of sound propagation in the ocean and principles of signal processing.

Event-Related Potentials (ERPs) were recorded from over the two cerebral hemispheres (occipital and parietal areas) during a 78 minute vigilance task. Twelve right-handed male subjects served in two experimental sessions. These subjects focussed on a central fixation point and responded to signals presented at unpredictable times in one of three locations: 2.5 deg. to right of the central fixation, central, and 2.5 deg to the left of center. Subjects decided whether to press a response key with either the left or right hand with each presentation. Signals were letter Ts oriented either right-side up or upside down. Six responded to T-up with right hands and six did the opposite. The ERPs were recorded from two scalp areas over left and right occipital cortex, and two sites above left and right parietal areas. Vertical and horizontal eye movements were also recorded. Two major ERP components emerged: a negative peak occurring at about 180 msec. after the stimulus, and a positive component appearing at about 320 msec. ERP latency asymmetries occurred at all recording sites as a function of visual field of stimulation. That is, for the N180 component stimuli presented in the left visual field (LVF) resulted in shorter latency ERPs at the contralateral (right) hemisphere than at the ipsilateral (left) hemisphere. A contralateral advantage was also observed for stimuli presented in the RVF.

The long term goal of this project is to develop efficient inversion algorithms for successful estimation and detection by incorporating (fully or partially) the physics of the propagation medium. Algorithms will be designed for robust ASW localization and detection and also for geoacoustic inversion.

The long term goal of this project is to develop efficient inversion algorithms for successful estimation and detection by incorporating (fully or partially) the physics of the propagation medium. Algorithms will be designed for robust ASW localization and detection and also for geoacoustic inversion.

The long term goal of this project is to develop efficient inversion algorithms for successful estimation and detection by incorporating (fully or partially) the physics of the propagation medium. Algorithms will be designed for robust ASW localization and detection and also for marine mammal localization and tracking.

The long term goal of this project is to develop efficient inversion algorithms for successful estimation and detection by incorporating (fully or partially) the physics of the propagation medium. Algorithms will be designed for robust ASW localization and detection and also for marine mammal localization and tracking.

An extended Prony algorithm applicable to signal processing techniques for clutter filtering and windshear detection is discussed. The algorithm is based upon modelling the radar return as a time series, and appears to offer potential for improving hazard factor estimates in the presence of strong clutter returns.

This thesis investigates the periodic autocorrelation function (PACF) and periodic ambiguity function (PAF) for orthogonal continuous waveform (CW) modulations used in netted low probability of intercept (LPI) radar. Three orthogonal polyphase sequences and one frequency coding sequence are examined and their PACF and PAF characteristics are quantified. The Wigner-Ville distribution (WVD) and quadrature mirror filter bank (QMFB) time-frequency signal processing techniques and the cyclostationary bi-frequency technique (often used in non-cooperative intercept receivers) are used to detect the orthogonal CW signals and extract their parameters. The results shows that a combination of the techniques used were able to extract the basic signal parameters of bandwidth and code period from the polyphase waveforms and also the frequency hop slots and code length from the frequency coding sequence. The concept of using a swarm of unmanned aerial vehicles (UAV) is examined from the viewpoint of a coordinated group of netted intercept receivers in search of an LPI radar network.

The long term goal of this project is to develop efficient inversion algorithms for successful estimation and detection by incorporating (fully or partially) the physics of the propagation medium. Algorithms will be designed for robust ASW localization and detection and also for geoacoustic inversion.

This document describes the progress on the work performed for "Moving Target Detection and Motion Estimation in Foliage Using Along Track Monopulse Synthetic Aperture Radar Imaging," under Contract N00014-96-1-0586, and "Signal Subspace processing of Uncalibrated MTD-SARs," under Contract N00014-97-1-0966 for the Office of Naval Research for the period ending on 9/30/98. Our main goal at this stage of the investigation was to test the relative merits of the proposed signal subpace method with a realistic SAR/ISAR database. For this purpose, we utilized the ISAR data of an airborne commercial aircraft; the ISAR data were collected by the radar group at the Navy's SPAWAR System Center, San Diego. In the first stage of our study, we used the non-overlapping block-based implementation of the signal subspace algorithm for registering two ISAR images of the aircraft which were formed via two separate slow-time segments of the target flight path. We also used the overlapping block-based signal subspace processing to estimate the motion field of the ISAR images. The study with the realistic ISAR data revealed interesting properties of the ISAR signatures of the reflectors, cavities and surface structure for an airborne target.

The long term goal of this project is to develop efficient inversion algorithms for successful estimation and detection by incorporating (fully or partially) the physics of the propagation medium. Algorithms will be designed for robust ASW localization and detection and also for geoacoustic inversion.

This paper addresses the problem of uncoded multiple access (MA) joint detection for the case in which user signatures are linearly dependent. The linearly dependent scenario would occur when the number of users in a communication system is increased beyond the dimension of the signal space available for transmission. This over-saturating of the signal space can, in principle, be accomplished with minimal impact on system performance, assuming that optimal detection can he implemented. The optimal detector for the general over-saturated case has a complexity that is exponential in the number of users. To find a joint detector for the MA communication of K N users in N dimensional signal space, the authors impose a hierarchical cross-correlation structure on the user signature waveforms. This paper develops a tree processing procedure that takes advantage of this structure to give the optimal estimate with an extremely low computational complexity. The authors show this complexity to be (in typical cases) a low-order-polynomial in the number of users. This is an enormous savings in computations over the O(2 sup K) computations needed if the signatures did not exhibit any structure.

The long term goal of this project is to develop efficient inversion algorithms for successful estimation and detection by incorporating (fully or partially) the physics of the propagation medium. Algorithms will be designed for robust ASW localization and detection and also for geoacoustic inversion.

This research program is aimed at developing a broad suite of techniques in detection, estimation, and reconstruction from images and video, which will help to advance the state of the art in automatic target recognition.

In the area of narrowband signal processing, design rules are developed for optimum decimator and interpolator, a new efficient scheme using recursive filter for decimation/interpolation is proposed, and a novel approach to the computation of narrowband spectra is shown to yield substantial saving over conventional approaches. Results on the implementation of recursive filters with poles near the unit circle that produces significantly reduced roundoff error include a transformation technique, a scheme to modify the quantizer error spectrum, and a new computationally efficient low noise filter structure. In the area of nonclassical signal detection, several results were derived on nonparametric sequential procedures and on the quantization of signal for detection. In addition, a programmable charge transfer device filter is developed, several problems concerning ADPCM are investigated, results are obtained on FFT roundoff error including the prime factor algorithm, and an effective method of generating random sequences is studied.

The objective of this work was to develop methodologies that will allow the human analyst to be removed from the processing loop. It has been shown in a number of recent demonstrations that when the most skilled practitioners process geophysical data, select data chips for analysis, select features for classification, select one of a suite of classifiers, and manually tune the classifier boundaries, excellent classification performance can be achieved. Here, we aim to develop techniques to improve target characterization and reduce classifier sensitivity to imprecision in the target characterizations, thereby reducing the need for an expert human analyst.

The long term goal of this project is to develop efficient inversion algorithms for successful estimation and detection by incorporating (fully or partially) the physics of the propagation medium. Algorithms will be designed for geoacoustic inversion and also for robust ASW localization and detection. Objectives: * Achieve accurate and computationally efficient source localization by designing estimation schemes that combine acoustic field modeling and optimization approaches. * Develop methods for passive localization and inversion of environmental parameters that select features of propagation that are essential to model for accurate inversion.

This report discusses a Broadband Bionic Sonar System and a signal processing technique for detection and identification of underwater targets. Since the sonar and radar target echoes are the solutions to the same Helmholtz Equation in their respective media, an Ultra Wideband Radar System can also be developed for target detection and identification base on the same principles and signal processing techniques. In the bionic sonar system with the resonance detection technique for detection and identification of underwater targets, it appears to mimic a dolphin's and/or a bat's acoustic sensory systems. It is conceivable that an Ultra-Wideband Radar System of the same detection technique can also be developed.

This thesis documents the use of the SRC-6 Reconfigurable Computer for use in analyzing low probability of intercept (LPI) signals using the Choi-Williams distribution. The SRC-6 is a reconfigurable computer manufactured by SRC Computers, Inc. which allows the user to tailor both the software and the hardware to a specific task. This increases the speed at which the task can be accomplished making it useful for applications in electronic intelligence (ELINT). The Choi-Williams distribution is a mathematical technique that was first created using MATLAB and then converted to C code for use on the SRC-6. The purpose of this study is to investigate the feasibility of using a reconfigurable computer for ELINT applications and the timely detection and classification of LPI signals. This thesis is part of a larger study to use reconfigurable computers for the autonomous detection and classification of LPI signals.

Using current technologies, the cost of identifying and disposing of UXO in the United States is estimated to range up to $500 billion. Site specific costs range from $400/acre for surface UXO to $1.4 million/acre for subsurface UXO. There are 1900 Formerly Used Defense Sites (FUDS) and 130 Base Realignment and Closure (BRAC) installations that need to be cleared. Several sensor modalities are currently being explored for the detection and identification of surface and buried UXO. These include electromagnetic induction (EMI), magnetometers, radar, and seismic sensors. These sensors experience little difficulty detecting the UXO, thus detection does not create the bottleneck that results in the high cost of remediating sites. The primary contributor to the costs and time associated with remediating a UXO contaminated site is the high false-alarm rate associated with each of the sensors when operated individually. In this project, we investigated the phenomenological aspects of the UXO detection, location, and discrimination problem using EMI, radar, and magnetometer sensors. The fundamental insight garnered by characterizing the underlying physics was transitioned into performance bounds, as well as high-performance sensor fusion and signal processing algorithms for enhanced detection, location, and discrimination of buried UXO under a wide range of environmental conditions. The signal processing algorithms that were developed were evaluated on data collected in the field during two government-sponsored sensor demonstrations. Finally, the algorithms were applied to a blind data set and performance was assessed by an independent agency.

The long term goal of this project is to develop efficient inversion algorithms for successful estimation and detection by incorporating (fully or partially) the physics of the propagation medium. Algorithms will be designed for robust ASW localization and detection and also for geoacoustic inversion.

The long term goal of this project is to develop efficient inversion algorithms for successful estimation and detection by incorporating (fully or partially) the physics of the propagation medium. Algorithms will be designed for geoacoustic inversion and also for robust ASW localization and detection.

High resolution windspeed profile measurements are needed to provide reliable detection of hazardous low altitude windshear with an airborne pulse Doppler radar. The system phase noise in a Doppler weather radar may degrade the spectrum moment estimation quality and the clutter cancellation capability which are important in windshear detection. Also the bias due to weather return Doppler spectrum skewness may cause large errors in pulse pair spectral parameter estimates. These effects are analyzed for the improvement of an airborne Doppler weather radar signal processing design. A method is presented for the direct measurement of windspeed gradient using low pulse repetition frequency (PRF) radar. This spatial gradient is essential in obtaining the windshear hazard index. As an alternative, the modified Prony method is suggested as a spectrum mode estimator for both the clutter and weather signal. Estimation of Doppler spectrum modes may provide the desired windshear hazard information without the need of any preliminary processing requirement such as clutter filtering. The results obtained by processing a NASA simulation model output support consideration of mode identification as one component of a windshear detection algorithm.

Signal subspace processing issues associated with uncalibrated along-track SARs for Moving Target Detection (MTD) and motion estimation are examined. A non-overlapping block-based implementation of the signal subspace algorithm for MTD and ATR is developed. Overlapping block-based signal subspace processing to estimate target motion parameters is studied. Application of the signal subspace method in other electronic and camera-based imaging systems, for example, diagnostic medicine and video, are investigated. Calibrating wide-beamwidth monopulse SARs is studied. Numerical results are provided.

Signal subspace processing issues associated with uncalibrated along-track SARs for Moving Target Detection (MTD) and motion estimation are examined. A non-overlapping block-based implementation of the signal subspace algorithm for MTD and ATR is developed. Overlapping block-based signal subspace processing to estimate target motion parameters is studied. Application of the signal subspace method in other electronic and camera-based imaging systems, for example, diagnostic medicine and video, are investigated. Calibrating wide-beamwidth monopulse SARs is studied. Numerical results are provided.

The long-term goal of this project is to develop methods for geoacoustic inversion, localization, and detection by combining the physics of sound propagation in the ocean and principles of signal processing.

Communications receivers are commonly optimized with respect to a stationary interference source. When the interference originates in a composite source, the optimal receiver may be represented as a collection of generically similar subreceivers each one of which is optimized with respect to one of the interference subsources. Selection of the correct subreceiver is conditioned on identifying the subsource that is interfering at a given time. This selection process is controlled by a statistical demultiplexer, In the absence of real communications, when the statistical demultiplexer operates on uncorrelated noise, its function is to estimate the present state of a spectrally white, non-Gaussian, nonstationary random process. To evaluate the performance of the statistical demultiplexer, it is necessary to assert that the state of the observed process is itself a stationary (unobservable) process. Consideration is given to the case in which the unobservable is Markov. When the receiver is an isolated system that draws power from a limited reservior of energy, and messages arrive only sporadically, the statistical demultiplexer in its simplest form detects the mere presence of the message and control a switch in the power line to the main receiver subsystem. The alertness strategy embodied in this scheme can greatly extend the life-time of the system under conditions that are considered in some detail.

The objectives of this study are to characterize apparent background motion as a function of observer-satellite orbit and sighting geometry and then to compare the effectiveness of signal processing filters in minimizing background motion effects on target detection. For many applications, complex algorithms cannot be realized economically, so simple yet effective algorithms must be developed to do the job. In this study both ideal and optimum third-order temporal recursive filters were synthesized and their simulated signal-to-noise ratio performance compared with that obtained by simple temporal differencing (first-through fourth-order). Our results indicate that for the same amount of memory, compared to the differencing filters, significant SNR performance improvement can be achieved with optimum recursive filters at the cost of a modest increase in filter complexity. The study is particularly pertinent to the detection by a staring mosaic sensor of aircraft from satellites in high (1000 kilometer) orbits. They are typically dim targets flying close to a structured background (the earth surface), and the background clutter tends to dominate the detection problem. (Author)

This thesis describes the implementation of an ELINT algorithm for the detection and classification of Low Probability of Intercept (LPI) signals. The algorithm was coded in the C programming language and executed on a Field Programmable Gate Array based reconfigurable computer; the SRC-6 manufactured by SRC Computers, Inc. Specifically, this thesis focuses on the preprocessing stage of an LPI signal processing algorithm. This stage receives a detected signal that has been run through a Quadrature Mirror Filter Bank and outputs the preprocessed signal for classification by a neural network. A major value of this study comes from comparing the performance of the reconfigurable computer to that of supercomputers and embedded systems that are currently used to solve the signal processing needs of the United States Navy.

This thesis describes the implementation of an ELINT algorithm for the detection and classification of Low Probability of Intercept (LPI) signals. The algorithm was coded in the C programming language and executed on a Field Programmable Gate Array based reconfigurable computer; the SRC-6 manufactured by SRC Computers, Inc. Specifically, this thesis focuses on the preprocessing stage of an LPI signal processing algorithm. This stage receives a detected signal that has been run through a Quadrature Mirror Filter Bank and outputs the preprocessed signal for classification by a neural network. A major value of this study comes from comparing the performance of the reconfigurable computer to that of supercomputers and embedded systems that are currently used to solve the signal processing needs of the United States Navy.

This thesis describes the implementation of an ELINT algorithm for the detection and classification of Low Probability of Intercept (LPI) signals. The algorithm was coded in the C programming language and executed on a Field Programmable Gate Array based reconfigurable computer; the SRC-6 manufactured by SRC Computers, Inc. Specifically, this thesis focuses on the preprocessing stage of an LPI signal processing algorithm. This stage receives a detected signal that has been run through a Quadrature Mirror Filter Bank and outputs the preprocessed signal for classification by a neural network. A major value of this study comes from comparing the performance of the reconfigurable computer to that of supercomputers and embedded systems that are currently used to solve the signal processing needs of the United States Navy.

The Wavelet in Signal Detection and Identification: Comparative Signal Processing Technology Evaluation Program has been conducted by Lockheed Missiles & Space Company to develop wavelet based signal detection and classification techniques and compare them to Fourier time-frequency methods. The problem domain is submarine detection and identification using transient passive sonar signals. These nontraditional signals are critical in solving current Anti-Submarine Warfare (ASW) problems which include ultra quiet nuclear submarines and third world diesel submarines operating on batteries in shallow water. The Wavelets program supports the ARPA sponsored Multisensor Fusion System Program (MSF) at Lockheed (ARPA contract N00014-91-C-0237).

This goal of this project was to develop novel schemes for the fusion of heterogeneous information. The target application was the detection and classification of personnel activity in both indoor and outdoor environments under dependent observations. We have identified features and designed a classifier that achieves up to 95% classification accuracy on classifying the occupancy with indoor footstep data. MDL-based copula selection strategies are investigated and a detector based on vines is designed that extends previous bivariate copula-based detectors to a multi-sensor application. Our copula-based detectors yield more than 40% improvement over the conventional data fusion methodologies. We extend our solution to quantized sensor information and demonstrate that injecting controlled noise can dramatically reduce computational complexity with insignificant performance loss. We propose and derive the Conditional Posterior Cram r-Rao Lower Bound (CPCRLB) for online tracking. We demonstrate that the PCRLB-based iterative approach converges quickly with significantly reduced computational cost as compared to a one-shot approach. Detector design in the presence of security threats, such as data falsification attacks to sensor networks, are also addressed. Error-control codes and decoding algorithms are used to reliably classify data in a network containing human agents.

This thesis investigates the periodic autocorrelation function (PACF) and periodic ambiguity function (PAF) for orthogonal continuous waveform (CW) modulations used in netted low probability of intercept (LPI) radar. Three orthogonal polyphase sequences and one frequency coding sequence are examined and their PACF and PAF characteristics are quantified. The Wigner-Ville distribution (WVD) and quadrature mirror filter bank (QMFB) timefrequency signal processing techniques and the cyclostationary bi-frequency technique (often used in non-cooperative intercept receivers) are used to detect the orthogonal CW signals and extract their parameters. The results shows that a combination of the techniques used were able to extract the basic signal parameters of bandwidth and code period from the polyphase waveforms and also the frequency hop slots and code length from the frequency coding sequence. The concept of using a swarm of unmanned aerial vehicles (UAV) is examined from the viewpoint of a coordinated group of netted intercept receivers in search of an LPI radar network.

This thesis investigates the periodic autocorrelation function (PACF) and periodic ambiguity function (PAF) for orthogonal continuous waveform (CW) modulations used in netted low probability of intercept (LPI) radar. Three orthogonal polyphase sequences and one frequency coding sequence are examined and their PACF and PAF characteristics are quantified. The Wigner-Ville distribution (WVD) and quadrature mirror filter bank (QMFB) timefrequency signal processing techniques and the cyclostationary bi-frequency technique (often used in non-cooperative intercept receivers) are used to detect the orthogonal CW signals and extract their parameters. The results shows that a combination of the techniques used were able to extract the basic signal parameters of bandwidth and code period from the polyphase waveforms and also the frequency hop slots and code length from the frequency coding sequence. The concept of using a swarm of unmanned aerial vehicles (UAV) is examined from the viewpoint of a coordinated group of netted intercept receivers in search of an LPI radar network.

This thesis documents the use of the SRC-6 Reconfigurable Computer for use in analyzing low probability of intercept (LPI) signals using the Choi-Williams distribution. The SRC-6 is a reconfigurable computer manufactured by SRC Computers, Inc. which allows the user to tailor both the software and the hardware to a specific task. This increases the speed at which the task can be accomplished making it useful for applications in electronic intelligence (ELINT). The Choi-Williams distribution is a mathematical technique that was first created using MATLAB and then converted to C code for use on the SRC-6. The purpose of this study is to investigate the feasibility of using a reconfigurable computer for ELINT applications and the timely detection and classification of LPI signals. This thesis is part of a larger study to use reconfigurable computers for the autonomous detection and classification of LPI signals.

The long term goal of this project is to develop efficient inversion algorithms for successful estimation and detection by incorporating (fully or partially) the physics of the propagation medium. Algorithms will be designed for robust ASW localization and detection and also for marine mammal localization and tracking. The objectives are to: 1. Achieve accurate and computationally efficient source localization by designing estimation schemes that combine full field modeling and search optimization. 2. Develop methods for passive localization and inversion of environmental parameters that select features of propagation that are essential to model for accurate inversion.

In signal processing applications it is common to assume Gaussian statistics in the design of optimal signal processors. However, non-Gaussian processes do arise in many situations. For example, measurements reveal that radar clutter may be approximated by either Weibull, K-distributed, Lognormal, or Gaussian distributions depending upon the scenario. When the possibility of a non-Gaussian problem is encountered, the question as to which probability distributions should be utilized in a specific situation for modeling the data needs to be answered. In practice, the underlying probability distributions are not known a priori. Consequently, an assessment must be made by monitoring the environment. Another consideration is that radar detection problems can usually be divided into strong, intermediate, and weak signal cases. Hence, the system that monitors a radar environment must be able to subdivide the surveillance volume into weak background noise and clutter patches in addition to approximating the underlying probability distributions for each patch. This is in contrast to current practice where a single robust detector, usually based on the Gaussian assumption, is employed. The objective of this work is to develop techniques that monitor the environment and select the appropriate detector for processing the data. The main contributions are: (1) an image processing technique is devised which enables partitioning of the surveillance volume into background noise and clutter patches, (2) the Ozturk algorithm is used to identify suitable approximations to the probability density function for each clutter patch, and (3) rules to be used with an expert system shell under development at the University of Massachusetts and Boston University are formulated for monitoring the environment and selecting the appropriate detector for processing the data.

In signal processing applications it is common to assume Gaussian statistics in the design of optimal signal processors. However, non-Gaussian processes do arise in many situations. For example, measurements reveal that radar clutter may be approximated by either Weibull, K-distributed, Lognormal, or Gaussian distributions depending upon the scenario. When the possibility of a non-Gaussian problem is encountered, the question as to which probability distributions should be utilized in a specific situation for modeling the data needs to be answered. In practice, the underlying probability distributions are not known a priori. Consequently, an assessment must be made by monitoring the environment. Another consideration is that radar detection problems can usually be divided into strong, intermediate, and weak signal cases. Hence, the system that monitors a radar environment must be able to subdivide the surveillance volume into weak background noise and clutter patches in addition to approximating the underlying probability distributions for each patch. This is in contrast to current practice where a single robust detector, usually based on the Gaussian assumption, is employed. The objective of this work is to develop techniques that monitor the environment and select the appropriate detector for processing the data. The main contributions are: (1) an image processing technique is devised which enables partitioning of the surveillance volume into background noise and clutter patches, (2) the Ozturk algorithm is used to identify suitable approximations to the probability density function for each clutter patch, and (3) rules to be used with an expert system shell under development at the University of Massachusetts and Boston University are formulated for monitoring the environment and selecting the appropriate detector for processing the data.

This article is from Sensors (Basel, Switzerland) , volume 14 . Abstract This study presents a fault detection of roller bearings through signal processing and optimization techniques. After the occurrence of scratch-type defects on the inner race of bearings, variations of kurtosis values are investigated in terms of two different data processing techniques: minimum entropy deconvolution (MED), and the Teager-Kaiser Energy Operator (TKEO). MED and the TKEO are employed to qualitatively enhance the discrimination of defect-induced repeating peaks on bearing vibration data with measurement noise. Given the perspective of the execution sequence of MED and the TKEO, the study found that the kurtosis sensitivity towards a defect on bearings could be highly improved. Also, the vibration signal from both healthy and damaged bearings is decomposed into multiple intrinsic mode functions (IMFs), through empirical mode decomposition (EMD). The weight vectors of IMFs become design variables for a genetic algorithm (GA). The weights of each IMF can be optimized through the genetic algorithm, to enhance the sensitivity of kurtosis on damaged bearing signals. Experimental results show that the EMD-GA approach successfully improved the resolution of detectability between a roller bearing with defect, and an intact system.

This article is from Sensors (Basel, Switzerland) , volume 14 . Abstract This study presents a fault detection of roller bearings through signal processing and optimization techniques. After the occurrence of scratch-type defects on the inner race of bearings, variations of kurtosis values are investigated in terms of two different data processing techniques: minimum entropy deconvolution (MED), and the Teager-Kaiser Energy Operator (TKEO). MED and the TKEO are employed to qualitatively enhance the discrimination of defect-induced repeating peaks on bearing vibration data with measurement noise. Given the perspective of the execution sequence of MED and the TKEO, the study found that the kurtosis sensitivity towards a defect on bearings could be highly improved. Also, the vibration signal from both healthy and damaged bearings is decomposed into multiple intrinsic mode functions (IMFs), through empirical mode decomposition (EMD). The weight vectors of IMFs become design variables for a genetic algorithm (GA). The weights of each IMF can be optimized through the genetic algorithm, to enhance the sensitivity of kurtosis on damaged bearing signals. Experimental results show that the EMD-GA approach successfully improved the resolution of detectability between a roller bearing with defect, and an intact system.