Loren W. Nolte

Optimal Detection and Performance of Distributed Sensor Systems

Global optimization of a distributed sensor detection system withfusion is considered, where the fusion rule and local detectors aresolved to obtain overall optimal performance. This yields coupledequations for the local detectors and the fusion center.The detection performance of the distributed system with fusionis developed. The globally optimal system performance is comparedwith two suboptimal systems. Receiver operating characteristics(ROCs) are computed numerically for the problem of detecting aknown signal embedded in non-Gaussian noise.

Active microwave imaging. I. 2-D forward and inverse scattering methods

Active microwave imaging (MWI) for the detection of breast tumors is an emerging technique to complement existing X-ray mammography. The potential advantages of MWI arise mainly from the high contrast of electrical properties between tumors and normal breast tissue. However, this high contrast also increases the difficulty of forming an accurate image because of increased multiple scattering. To address this issue, we develop fast forward methods based on the combination of the extended Born approximation, conjugate- and biconjugate-gradient methods, and the fast Fourier transform. We propose two nonlinear MWI algorithms to improve the resolution for the high-contrast media encountered in microwave breast-tumor detection. Numerical results show that our algorithms can accurately model and invert for the high-contrast media in breast tissue. The outcome of the inversion algorithms is a high-resolution digital image containing the physical properties of the tissue and potential tumors.

A posteriori probability source localization in an uncertain sound speed, deep ocean environment

The a posteriori source position probability density function for a narrow-band source in an uncertain acoustic environment is derived. The implementation of this probability density function (pdf) is termed the optimum uncertain field processor (OUFP). It is shown that the OUFP is a generalization of matched-field processing to situations in which there is uncertainty about the environment. The robustness of the OUFP is illustrated through performance comparisons to a matched-field algorithm.

Design and Performance Comparison of Distributed Detection Networks

Optimization of a distributed detection network using theminimum global cost criterion results in local processors thatindividually form the likelihood ratio when the input observationvectors are statistically independent. In addition, the localthresholds and the network performance can be expressed as afunction of the receiver operating characteristics (ROCs) of the localprocessors. The performance of rive distributed networks arecompared numerically using local ROCs from the conic ROCfamily.

More on the Detection of One of M Orthogonal Signals

Receiver operating characteristics (ROCs) for the classic problem of detecting the presence or absence of one of M orthogonal signals is presented. Previous results were valid for low detectability, for which the ROC is approximately normal (i.e., appears as a straight line with unit slope on normal normal probability paper) and the detectability depends on the logarithm of the number of possible signals M. For high detectability, however, the ROC departs from normality. In addition, the rate at which detectability decreases as M increases is more rapid than that predicted by the classical approximation.

COMPUTATIONALLY EFFICIENT MONTE CARLO ESTIMATION ALGORITHMS FOR MATCHED FIELD PROCESSING IN UNCERTAIN OCEAN ENVIRONMENTS

In this paper, Monte Carlo estimation techniques are presented for computationally efficient implementation of two methods for matched field source localization in uncertain ocean channels. In the Optimal Uncertain Field Processor (OUFP), Monte Carlo integration is used to integrate out the environmental parameters and thus estimate the a posteriori distribution of the source location parameters. In the Minimum Variance Beamformer with Environmental Perturbation Constraints (MV-EPC), Monte Carlo estimation of the signal correlation matrix averaged over the ensemble of environmental realizations is used to estimate the beamformer constraints. Using the OUFP, detection performance bounds are evaluated which indicate that source position uncertainty affects performance much more than environmental uncertainty. An upper bound on source localization performance is also obtained indicating that for short observation times a threshold signal-to-noise ratio (SNR) exists, dependent upon environmental uncertainty, below which source localization performance rapidly degrades. Among robust minimum variance beamforming methods, the MV-EPC method demonstrated superior probability of correct localization (PCL), both in single source scenarios and in the presence of interference. The OUFP at high SNR and the MV-EPC at large observation times both achieved near perfect source localization performance, although for large environmental uncertainty the OUFP provides an upper bound on PCL.

Performance of the hough transform and its relationship to statistical signal detection theory

Abstract A statistical signal detection theory approach can be used for curve detection in digital images corrupted by random noise. This approach is shown to be a refinement of the frequently used Hough transform, resulting in improved performance, both in deciding the presence or absence of a curve in the image and in determining the location in the image of an existing curve. Detection performance is evaluated using receiver operating characteristics (ROC) curves. Location estimation performance is measured by deriving equations for both the Hough transform and the signal detection theory approach for the probability of correctly estimating the location of a curve in noise. The performances of these two approaches are compared for various signal-to-noise ratios and found to be significantly different for some values of signal-to-noise ratio.

An Anthropomorphic Breast Model for Breast Imaging Simulation and Optimization

RATIONALE AND OBJECTIVES Optimization studies for x-ray-based breast imaging systems using computer simulation can greatly benefit from a phantom capable of modeling varying anatomical variability across different patients. This study aimed to develop a three-dimensional phantom model with realistic and randomizable anatomical features. MATERIALS AND METHODS A voxelized breast model was developed consisting of an outer layer of skin and subcutaneous fat, a mixture of glandular and adipose, stochastically generated ductal trees, masses, and microcalcifications. Randomized realization of the breast morphology provided a range of patient models. Compression models were included to represent the breast under various compression levels along different orientations. A Monte Carlo (MC) simulation code was adapted to simulate x-ray based imaging systems for the breast phantom. Simulated projections of the phantom at different angles were generated and reconstructed with iterative methods, simulating mammography, breast tomosynthesis, and computed tomography (CT) systems. Phantom dose maps were further generated for dosimetric evaluation. RESULTS Region of interest comparisons of simulated and real mammograms showed strong similarities in terms of appearance and features. Noise-power spectra of simulated mammographic images demonstrated that the phantom provided target properties for anatomical backgrounds. Reconstructed tomosynthesis and CT images and dose maps provided corresponding data from a single breast enabling optimization studies. Dosimetry result provided insight into the dose distribution difference between modalities and compression levels. CONCLUSION The anthropomorphic breast phantom, combined with the MC simulation platform, generated a realistic model for a breast imaging system. The developed platform is expected to provide a versatile and powerful framework for optimizing volumetric breast imaging systems.

Wideband optimal a posteriori probability source localization in an uncertain shallow ocean environment

A wideband extension of the Optimum Uncertain Field Processor (OUFP) is presented. Combined with Monte Carlo estimation methods, this processor provides a fast, efficient, and robust technique for Matched-Field Processing (MFP). Under a simulated Hudson Canyon environment, a quantitative, probabilistic analysis of the sensitivity of the Optimum Matched-Field Processor (OMFP) to various kinds of environmental mismatch is shown. Similarly, the performance of the OUFP is calculated. Finally, the optimum wideband OUFP is applied to a data set collected from the Hudson Canyon, the results tabulated, and then compared to a standard Bartlett processor. The performance of the optimum wideband OUFP and the suboptimum Bartlett processor are in very good agreement with the performance predicted from simulation results.

Covariance between Fourier coefficients representing the time waveforms observed from an array of sensors

Typically, the assumption of a long observation interval is made to achieve uncorrelatedness between Fourier coefficients at different frequency indicies. An explicit relationship between observation interval length and covariance between the Fourier coefficients is obtained to provide insight as to how long the observation interval should be to approximate this desirable condition.Subject Classification: [43]60.20.