Nonlinear Detector Research Articles

Unsupervised iterative detection of land mines in highly cluttered environments

An unsupervised iterative scheme is proposed for land mine detection in heavily cluttered scenes. This scheme is based on iterating hybrid multispectral filters that consist of a decorrelating linear transform coupled with a nonlinear morphological detector. Detections extracted from the first pass are used to improve results in subsequent iterations. The procedure stops after a predetermined number of iterations. The proposed scheme addresses several weaknesses associated with previous adaptations of morphological approaches to land mine detection. Improvement in detection performance, robustness with respect to clutter inhomogeneities, a completely unsupervised operation, and computational efficiency are the main highlights of the method. Experimental results reveal excellent performance.

Dynamic range of fluorescence detection and base‐calling accuracy in DNA sequencer based on single‐photon counting

Recently, we developed a family of high-performance automated capillary DNA sequencing instruments based on a single-photon detection of fluorescently labeled DNA fragments. Our machines employ digital and broadband techniques, essential for achieving superior instrument sensitivity and dynamic range. In the present paper, we discuss limitations of the instrument's performance caused by the nonlinearity of single-photon detectors as well as methods for nonlinearity compensation which increase the detection dynamic range and base-calling accuracy.

Design of detectors based on stochastic resonance

This paper presents a study of the phenomenon of stochastic resonance in quantizers, and discusses the use of this phenomenon for the detection of weak sinusoidal signals in noise. Stochastic resonance in 2-level, symmetric 3-level, and symmetric multilevel quantizers is investigated. Expressions are derived for the signal-to-noise ratio (SNR) gain of the quantizers driven by a small amplitude sinsuoidal signal and i.i.d. noise. The gain depends on the probability density function (PDF) of the input noise, and for a given noise PDF, the gain can be maximized by a proper choice of the quantizer thresholds. The maximum gain G SR is less than unity if the input noise is Gaussian, but several non-Gaussian noise PDFs yield values of G SR exceeding unity. Thus, the quantizers provide an effective enhancement in the SNR, which can be utilized to design a nonlinear signal detector whose performance is better than that of the matched filter. The nonlinear detector in consideration consists of a stochastically resonating (SR) quantizer followed by a correlator. An asymptotic expression for the probability of detection of the SR detector is derived. It is shown that the detection performance of the SR detector is better than that of the matched filter for a large class of noise distributions belonging to the generalized Gaussian and the mixture-of-Gaussian families.

Improved calibration of the spectral responsivity of interference filter radiometers in the visible and near infrared spectral range at PTB

At the Physikalisch-Technische Bundesanstalt (PTB) silicon diode-based narrow-band interference filter radiometers (FR) are used for the radiometric determination of thermodynamic temperatures. To extend the temperature range down to 400 °C, the centre wavelength of the FRs was shifted from the visible to the near infrared (900 nm, 1000 nm and 1595 nm). An improved calibration procedure is presented which achieves relative uncertainties for the spectral responsivity of the FRs of the order of 10−4, by taking into consideration the increased temperature coefficient of the spectral responsivity and the nonlinearity of the silicon detectors when operated in the near infrared wavelength region.

Using Numerical Solutions in the Fourier Method for the Fokker-Planck-Kolmogorov Equation in the Analysis of First-Order Stochastic Nonlinear Systems for Nonlinear Detectors

We present a method for analyzing the characteristics of nonlinear detectors using the algorithms of first-order nonlinear differential equations. This method is based on numerical solutions of the Fokker–Planck–Kolmogorov (FPK) equations in the form of series of functions over Hermite–Chebyshev polynomials for both nonlinear systems and their linear counterparts. The results of the solutions for the linear case are extended to nonlinear systems in a recurrent way.

Adaptive loading for OFDM/SDMA-based wireless networks

The two major obstacles toward high-capacity indoor wireless networks are distortion due to the indoor channel and the limited availability of bandwidth which necessitates a high spectral efficiency. A combined orthogonal-frequency division multiplexing/spatial-diversity multiple access (OFDM/SDMA) approach can effectively tackle both obstacles. The channel distortion due to multipath propagation is easily mitigated with OFDM while the bandwidth efficiency can be increased with the use of SDMA. In order to keep the network's cost acceptable, simplified SDMA processors are preferred over the exponentially complex optimal maximum-likelihood processors. However, these simplified processors perform significantly worse than the optimal ones in terms of average bit error rate (BER). In this paper, we show that by adapting the constellation sizes applied on the individual subcarriers to the channel conditions, the performance of OFDM/SDMA processors can be significantly enhanced. In the uplink, we derive simple closed-form equations for the optimal constellation sizes for both a simple linear minimum mean-square error (MMSE) detector and for a nonlinear MMSE decision feedback equalizer (DFE) detector. Furthermore, for each detector, we introduce a simplified loading algorithm which lowers the computational and signaling complexity substantially at a small performance penalty. In the downlink, we study the dual precoders of the uplink detectors, respectively, the linear MMSE precoder and the nonlinear Tomlinson-Harashima (TH) MMSE-based precoder. For both precoders, we derive expressions for the optimal and the simplified constellation sizes. Additionally, we show that in time-division duplexing systems, the constellation distribution of a set of dual detectors/precoders is identical for up- and downlink, which effectively halves the computational complexity of adaptive loading. In the fully loaded uplink, the proposed adaptive loading algorithm results in a gain of 9 dB for a BER=10/sup -3/ for the linear MMSE detector and a gain of 4.5 dB for the nonlinear MMSE-DFE detector. In the fully loaded downlink, a gain of 6.3 dB is achieved for the MMSE precoder and 5.5 dB for the TH-MMSE precoder.

Multi-user detection in multicarrier-CDMA systems

The upcoming third generation mobile radio system in Europe is based on UMTS (Universal Mobile Telecommunications Standard). In order to supply access to a common transmission channel for several users, UMTS incorporates Code Division Multiple Access (CDMA). Besides a lot of practical advantages, CDMA suffers from multi-user interference limiting spectral efficiency dramatically. However, bandwidth is a very valuable resource and should be used as efficiently as possible. One appropriate mean to increase spectral efficiency of CDMA systems is multi-user detection. This paper gives an overview of different multi-user detection techniques and compares them with the conventional single-user detection including channel coding. Specifically, linear as well as nonlinear multi-user detectors are considered. Efficient realizations of linear detectors are presented and extended to improved nonlinear techniques. It is shown that nonlinear MUD including channel decoding can achieve a spectral efficiency twice as high as that of the well-known GSM standard (Global System for Mobile Communications) employing TDMA and FDMA.

Extinction Coefficient Determination of Polymeric Materials and Their Use in the Optical Characterization of Infrared Microscopes: A Preliminary Investigation into Quantitative Infrared Microspectroscopy

A comparison of infrared spectra collected in both a macro sample configuration and on a microscope accessory was conducted. The results of this comparison demonstrated that as the microscope aperture was reduced, absorbance values increased for a given polymer film by as much as 33% as compared to the same data collected in the corresponding macro experiment. A detailed evaluation of the optical pathlength for both the micro and macro configurations was conducted. Results of this evaluation and that of detector non-linearity showed that the latter effect was most problematic. During the course of the investigation the average angle of light traversing through the objective was determined for several infrared microscopes. Further analysis indicated that changing the aperture size of the microscope can have significant influence on the absorbance values obtained. In addition, extinction coefficients of poly(methyl methacrylate), polystyrene, poly(vinyl acetate), polycarbonate, and poly(ethylene terephthalate) were determined using macro FT-IR spectroscopy.

Extended depth of field with a nonlinear silver-halide emulsion detector.

Presented here are experimental results of an extended depth-of-field (EDF) system using film to record the image. The EDF system used is an application of wave-front coding. A phase mask is placed in the aperture stop of the lens to code the image data; then digital signal processing is used to decode the image. Wave-front coding is based on linear systems theory, but film is not a linear medium. An extra step of linearization is then needed to apply the technique. Both nonlinear and linearized results are shown, as is a test image that demonstrates the success of the technique for nonlinear media.

Signal detection theory, detectability and stochastic resonance effects.

Stochastic resonance is a phenomenon in which the performance of certain non-linear detectors can be enhanced by the addition of appropriate levels of random noise. Signal detection theory offers a powerful tool for analysing this type of system, through an ability to separate detection processes into reception and classification, with the former generally being linear and the latter always non-linear. Through appropriate measures of signal detectability it is possible to decide whether a local improvement in detection via stochastic resonance occurs due to the non-linear effects of the classification process. In this case, improvement of detection through the addition of noise can never improve detection beyond that of a corresponding adaptive system. Signal detection and stochastic resonance is investigated in several integrate-and-fire neuron models. It is demonstrated that the stochastic resonance observed in spiking models is caused by non-linear properties of the spike-generation process itself. The true detectability of the signal, as seen by the receiver part of the spiking neuron (the integrator part), decreases monotonically with input noise level for all signal and noise intensities.

Optimal Absorbance for Transmission or Reflection Spectra Measured under Conditions of Constant Detector Noise in the Presence of Stray Radiation

The signal-to-noise ratio ( S/N) for absorption spectrometers can be expressed as absorbance/absorbance noise ( Am/ SAm), where Am is the measured absorbance and Smm is the standard deviation in Am, or as concentration/concentration noise ( c/sc) calculated from Beer's law. For measurements made under ideal conditions, Am/ SAm. However, when stray radiation, or an analogous phenomenon, such as detector nonlinearity in FT-IR measurements or surface reflection in diffuse reflection spectrometry, contributes to the spectrum, C/ Sc ≠ Am/ SAm. It is known that the optimum value of Am/ SAm in the absence of stray light is found at an absorbance of 0.4816 when noise from the reference spectrum is taken into account and 0.4343 when noise from the reference spectrum is excluded from the calculations. In the present paper we discuss the effect of stray light on the absorbance and concentration S/N. The maximum value of c/sc is found at lower absorbance than is the case if stray light is absent. While this effect is negligibly small for the typical levels of stray light found with most contemporary monochromators, it can be significant when, for example, the response of a detector used in Fourier transform spectrometry is nonlinear or when the front-surface reflection in diffuse reflection measurements is large. For effective stray light levels of 10%, the maximum value of c/sc is found at an absorbance that is 22% lower than when no source of stray light is present.

Calibration in thermal field flow fractionation with polydisperse standards: Application to polyolefin characterization

A new procedure for calibrating ThFFF systems by using broad molecular weight standards based on a non-linear least squares fitting of the fractogram to the Molecular Weight Distribution to derive the system calibration parameters is presented and explored by simulation. This procedure proved robust with respect to noise and it is able to account for detector nonlinearity. The effect of band broadening was evaluated and it was found that significant bias is observed only for polydispersities lower than 1.3. The method was experimentally validated by using polystyrene in decalin and was then applied to polypropylene under conditions of high temperature ThFFF (Tc = 394 K). It appears that the presented method can be useful in characterising polyolefins polymers for which monodisperse standards are not available.

CDMA multiuser detection: a nonlinear programming approach

The optimum receiver to detect the bits of multiple code-division multiple access (CDMA) users has an exponential complexity in the number of active users in the system. Consequently, many suboptimum receivers have been developed to achieve good performance with less complexity. We take the approach of approximating the solution of the optimum multiuser detection problem (OMUD) using nonlinear programming relaxations. First, we observe that some popular suboptimum receivers indeed correspond to relaxations of the optimal detection problem. In particular, one proposed approximation method yields to iterative solutions which correspond to previously proposed heuristic nonlinear detectors. Using a nonlinear programming approach, we identify the convergence properties of these iterative detectors. Secondly, we propose a relaxation that yields a receiver which we call the generalized minimum mean squared error detector. We give a simple iterative implementation of the detector. Its performance is evaluated and comparisons to other suboptimum detection schemes are given.

Chromatographic alignment by warping and dynamic programming as a pre-processing tool for PARAFAC modelling of liquid chromatography–mass spectrometry data

Solutes analysed with LC–MS are characterised by their retention times and mass spectra, and quantified by the intensities measured. This highly selective information can be extracted by multiway modelling. However, for full use and interpretability it is necessary that the assumptions made for the model are valid. For PARAFAC modelling, the assumption is a trilinear data structure. With LC–MS, several factors, e.g. non-linear detector response and ionisation suppression may introduce deviations from trilinearity. The single largest problem, however, is the retention time shifts not related to the true sample variations. In this paper, a time warping algorithm for alignment of LC–MS data in the chromatographic direction has been examined. Several refinements have been implemented and the features are demonstrated for both simulated and real data. With moderate time shifts present in the data, pre-processing with this algorithm yields approximately trilinear data for which reasonable models can be made.

Fundamental effects and non-linear Si detector response

Non-linearity in the energy response of a Si p-i-n charged particle detector has been studied for incident particles with Z 1 between 3 and 26, and energies between 0.1 and 0.7 MeV per nucleon. Although the data closely followed a straight line relations, fitting of the data to a third order polynomial revealed that the response exhibited a persistent curvature that acted to reduce the energy interval spanned by a channel as the energy increased. The curvature increased as Z 1 increased from 4 to 8 and then systematically decreased. The curvature is larger and has the opposite energy dependence to the stopping in a dead entrance window and the energy deposited in non-ionising processes within the active layer. The plasma recombination dependence on the average stopping along the plasma column may account for the reduction in curvature as Z 1 increases from 9 to 25 but cannot explain the net effect. The low-energy increase in energy channel span, which has also been reported by others, might be associated with electron excitation in resonant and direct classical quasi-elastic collisions for low-energy ions, or less likely, electronic non-linearity's associated with Z 1 and energy dependent time structure in the current pulse from the detector. Simple interpolation of the window-loss corrected polynomial coefficients is the best approach if the calibration for Z 1 cannot be established directly.

Turbo Multiuser Detection for TD-CDMA

Summary In this paper the novel detection scheme multi-step joint detection for TD-CDMA mobile radio systems is presented. Multi-step joint detection uses the turbo principle for iteratively improving data detection. Extrinsic information obtained by FEC-decoding in a previous step is used for a joint reduction of interference and improvement of asymptotic efficiency of the linear multiuser detector. Multi-step joint detection helps to overcome the problems of small asymptotic efficiencies present in linear multiuser detectors like the zero forcing equalizer in TD-CDMA mobile radio systems in the case of high system loads. Simultaneously complexity is much lower than that of optimum nonlinear multiuser detectors based on the Viterbi algorithm. As an alternative to simulations the performance of multi-step joint detection can be theoretically determined under certain not too restrictive assumptions. Both approximate and simulative performance results are presented in the paper. It is shown that in typical mobile radio systems the required SNR at the receiver input can be reduced by approximately 10 dB compared to linear multiuser detection. Thus, multi-step joint detection helps to increase the permissible number of mobile stations per cell or to decrease the required cluster size in TD-CDMA mobile radio systems and thus improves spectrum efficiency and capacity.

Turbo decoding for robust multiuser detection in impulsive noise channels

A more robust version of the nonlinear M-decorrelating detector through iterative (turbo) decoding is proposed. The turbo code performance in the linear decorrelating detector is also studied. Simulation results indicate that combined nonlinear signal processing and turbo decoding can achieve highly robust performance in impulsive noise.

Accuracy and Reproducibility of X-ray Texture Measurements on Thin Films

AbstractRocking curve texture measurements were made on thin films of zinc oxide (ZnO) and platinum (Pt) using a powder x-ray diffractometer, and, in the case of ZnO, an area detector. The intensity corrections for defocussing and other geometric factors were made using a technique and associated software (Texture Plus*) developed at NIST. In both thin film systems, the texture was axisymmetric (fiber) and sharp, with full width at half maximum values of about 2.5°. Care was taken in the Pt case to ensure that the linear range of the x-ray detector was used to measure the intensities; for the ZnO data the degree of detector non-linearity was determined, and corrections were applied where necessary. The suitability of the Pt films for thin film texture standards was studied.

Propagation of Errors in the Precision and Accuracy of Quantitative Electron Microbeam Analysis

A common question, but one typically difficult to answer, is “How accurate is a microprobe or SEM analysis?” Commercial software for most EDS and WDS systems print out error limits that are typically calculated standard deviations using Gaussian counting statistics based on the accumulated peak and background counts. This is not a true measure of even the measurement precision, since random instrumental instabilities typically degrade “true” measurement precision (as determined by replicate measurements on standards) to 1.5 - 3+ times that predicted from Gaussian statistics. in addition to these random instrumental variations are a plethora of possible systematic and semirandom errors such as detector or faraday cup non-linearities, errors in estimation of background, EDS spectral deconvolution errors, WDS positional non-reproducibility, temperature and pressure variation effects on spectrometer efficiencies, high voltage drift or sample charging, surface contamination or oxidation, and sample roughness or tilt.

Practical measurement of femtosecond optical pulses using time-resolved optical gating

We report a practical experimental implementation of time-resolved optical gating (TROG) using a dispersive pulse (DP) propagation geometry and a two-photon absorption nonlinear detector. A measurement of 50 fs pulses from a self-mode-locked Ti:sapphire laser is demonstrated and a general procedure for calibrating the dispersion axis of the TROG trace is presented. A complete iterative pulse retrieval algorithm is described and the limitations of the DP-TROG technique are discussed.