Amount Of Additive Noise Research Articles

Maximum likelihood estimation for linearity testing of ADCs stimulated by known constant signals

A maximum likelihood (ML) estimator is derived for the problem of measuring the code transition levels of an analog-to-digital converter (ADC). The proposed method is intended to test the linearity of the ADC in the static regime, using only constant test signals, except for a small amount of additive noise. The measurement data are employed in a nearly optimal manner, due to the statistical properties of the ML estimator, which are thoroughly examined. The reported analysis allows the design of the test under a given uncertainty constraint.

CCD Color Camera Characterization for Image Measurements

In this article, we will analyze a range of different types of cameras for its use in measurements. We verify a general model of a charged coupled device camera using experiments. This model includes gain and offset, additive and multiplicative noise, and gamma correction. It is shown that for several cameras, except a typical consumer webcam, the general model holds. The associated model parameters are estimated. It is shown that for most cameras the model can be simplified under normal operating conditions by neglecting the dark current. We further show that the amount of additive noise is exceeded by the amount of multiplicative noise at intensity values larger than 10%-30% of the intensity range

Comments on ‘Is Partial Coherence a Viable Technique for Identifying Generators of Neural Oscillations?’

To aid prospective neural connectivity inference analysts and hoping to preclude misconception spread, we exploit the didatic value of some of the issues raised by Albo et al. (Biol Cybern 90: 318-326, 2004) who claim that signal-to-noise ratio (SNR) values can lead to mistakes in structural inference when using partial coherence in connection to Gersch's 1970 method for spotting signal sources (Gersch in Math Biosci 14: 177- 196, 1972). We show theoretically that Gersch's method is able only to spot which measurement of some common underlying factor has the least amount of additive noise and that this has nothing to do with any reasonable notion of 'causality' as suggested by Albo et al. (Biol Cybern 90: 318-326, 2004). We also show that despite the inherent structural ambiguity of the model used by Albo et al. (Biol Cybern 90: 318-326, 2004) to back their claim, its data can nonetheless furnish the correct time precedence hierarchy between the activities in its measured structures, both when simple (correlation) and more sophisticated methods are used (partial directed coherence) (Baccala and Sameshima in Biol Cybern 84:463-474, 2001a) in a true depiction of time series causality.

Blind identification and equalization of two-channel FIR systems in unbalanced noise environments

Blind identification is a very significant problem in many contexts where only the output of transmission channels can be observed. The solutions that can be found in the literature are limited to the case of equal amounts of additive noise on the observations. This paper proposes new identification procedures that can be applied to the case of two FIR channels affected by unknown and unbalanced amounts of additive noise. The identified models are then used for the minimal variance deconvolution of the unknown input signal. Several Monte Carlo simulations also confirm the good performance of these procedures in severe SNR conditions.

Combining Singular-Spectrum Analysis and neural networks for time series forecasting

In this paper, we propose a combination of an adaptive noise-reduction algorithm based on Singular-Spectrum Analysis (SSA) and a standard feedforward neural prediction model. We test the forecast skill of our method on some short real-world and computergenerated time series with different amounts of additive noise. The results show that our combined technique has better performances than those offered by the same network directly applied to raw data, and therefore is well suited to forecast short and noisy time series with an underlying deterministic data generating process (DGP).

Dynamics controlled by additive noise

Analysis is presented of a system, derived previously as a low-order model for the shear instability of convection, whose dynamics are dramatically simplified by tiny amounts of additive noise. The dynamics divide naturally into two phases. In the slower phase, trajectories are close to an invariant manifold; this allows small random disturbances to exert a controlling influence. A map is derived which provides an accurate description of the trajectories.

Effects of additive noise on four bar target detection as a function of contrast and spatial frequency

A psychovisual experiment with 80 observers determined the amount of additive noise an observer could tolerate in resolving 36 different bar targets. The results show that the detection threshold noise can be represented by a simple analytical expression with contrast and spatial frequency as variables. The expression that closely matched the data is Vnoise (6.4C + 0.68)/v' , where Vnoise IS the amount of noise that can be tolerated, C is the contrast, and is the spatial frequency in cycles! radian. The numerical values are a function of the monitor used and are readily obtained by a simple measurement.

Output residence time control

The problem of residence time control is extended to systems with outputs. Necessary and sufficient conditions for output residence time controllability in linear systems with small amounts of additive noise are derived. Design techniques for state-feedback controllers are developed and applied to a robotics control problem. The approach is based on an extension of the asymptotic first passage time theory to output processes. >

Limits of visual communication: the effect of signal-to-noise ratio on the intelligibility of American Sign Language.

To determine the limits of human observers' ability to identify visually presented American Sign Language (ASL), the contrast s and the amount of additive noise n in dynamic ASL images were varied independently. Contrast was tested over a 4:1 range; the rms signal-to-noise ratios (s/n) investigated were s/n = 1/4, 1/2, 1, and infinity (which is used to designate the original, uncontaminated images). Fourteen deaf subjects were tested with an intelligibility test composed of 85 isolated ASL signs, each 2-3 sec in length. For these ASL signs (64 x 96 pixels, 30 frames/sec), subjects' performance asymptotes between s/n = 0.5 and 1.0; further increases in s/n do not improve intelligibility. Intelligibility was found to depend only on s/n and not on contrast. A formulation in terms of logistic functions was proposed to derive intelligibility of ASL signs from s/n, sign familiarity, and sign difficulty. Familiarity (ignorance) is represented by additive signal-correlated noise; it represents the likelihood of a subject's knowing a particular ASL sign, and it adds to s/n. Difficulty is represented by a multiplicative difficulty coefficient; it represents the perceptual vulnerability of an ASL sign to noise and it adds to log(s/n).

Application of the viterbi algorithm to the detection of compact spectrum

A Viterbi detector is described-for detection of the ternary compact spectrum output waveform presented by Veillard [1]. The performance of the detector with additive noise is compared by simulation, to threshold detection of the same waveform. It is demonstrated that the implementation of the Viterbi algorithm for this case is relatively simple. The Viterbi detector presented is shown to produce a lower bit error rate than theshold detection for a given amount of additive noise. It is also shown that 4-bit quantization before Viterbi detection is sufficient to produce results essentially the same as those obtained without quantization.

Visual-error criterion for color-image processing

We extend a new visual-error criterion for achromatic images to color images. After the tristimulus planes are transformed to a luminance-compatible color space, the achromatic-error criterion is used for the luminance plane and the squared-error criterion is used for the two chromatic planes taken together. The optimal linear estimator for the resulting color-image-error criterion is compared against a purely squared-error criterion in the task of reconstructing color images degraded by varying amounts of additive noise. The images restored with the new error criterion appeared sharper and truer in fine detail. Except for the least and most noisy cases, in which no clear preference was found, viewers of the cathode-ray-tube display unanimously preferred these restorations over the comparatively fuzzy squared-error ones, despite a grainier appearance. These results are consistent with the previous ones found with the new visual-error criterion for achromatic images only. Through this extension to color images and the good results in restoring noisy images, we have found an error criterion that shows promise for applications in color-image processing.

Data Compression and Enhancement of Sampled Images

A two-dimensional sampled image can be represented by an equal number of spatial frequencies in the Fourier domain. However, due to physical limitations of the sampling process, there exists an effective cutoff frequency beyond which no information is preserved. The knowledge of this cutoff frequency is very important, since a considerable amount of additional noise can be introduced by frequency components above this cutoff. Availability of a sharp edge within the image allows, within the linear theory, the estimation of the transfer function of the digitizing process itself. This calculated transfer function may be used for image enhancement below the cutoff frequency. In addition, significant amount of data compression may be achieved by removing all spatial frequency terms above the cutoff frequency. An example of this technique is developed utilizing a tribar resolution chart sampled at 1024 x 1024 points.

Background Noise in Nonlinear Oscillators

For a general black box model of a resistive one terminal-pair nonlinear oscillator, it is shown that the noise output from noise bands around the oscillatory frequency is composed of an additive noise of the shape of the oscillator resonant circuit and a very small FM broadening of the oscillator line. A noise generator across the oscillator terminals will produce the same amount of additive noise in the load as if the nonlinearity were absent, but the upper and lower sidebands are fully correlated, not independent as in the linear case. The noise bandwidth of the additive noise can be determined by the passive elements and the change in oscillator output power with changing load conductance at the operating load conductance. When the load conductance is adjusted for maximum oscillator output power, the noise bandwidth is the same as if the value of load conductance gave maximum power from a linear circuit.