Fluctuation and Noise Letters

We investigate fixed dopants in the presence of mobile point defects like foreign atoms or vacancies. A mobile defect entering the first Bohr radius R_B of a dopant will modulate the generation-recombination process. The times a defect walks inside or outside R_B are shown to be power-law distributed giving rise to 1 / f^b noise. The predicted Hooge coefficient α_def depends on R_B, on the normalized fluctuations of charge carriers and on the number of charge carriers compared to lattice sites; our model suggests that the magnitude of 1/f noise can be decreased at will by increasing the ionization of dopants.

The influence of white and colored noise on the dynamics of self-sustained oscillator is considered in the regime of spiral (phase-coherent) chaos. The effective phase diffusion coefficient and power spectra are analyzed for noisy chaotic self-sustained oscillations. We show that chaotic oscillations can be synchronized by external narrow-band noise. Effects of chaos synchronization are compared for narrow-band noise signals with similar spectra but distinct probability distributions.

We demonstrate the use of the one-and-a-half spectrum as a way of looking at the time-asymmetry of Barkhausen noise. This method is of particular value because it allows us to look in the regime where individual Barkhausen pulses are no longer distinguishable. We use the one-and-a-half spectrum on Barkhausen data from two different ferromagnetic samples and find asymmetries in the noise of both. The magnitude of the asymmetry is different in the two samples and we find no clear change in the asymmetry when we move from the regime of well-separated jumps at low field rate to the continuous fluctuating signal at higher rates.

We characterize granular NbN_x thin cermet films deposited on either sapphire substrate or on SiO₂ and compare the 1/f noise at 300 K and 80 K. The films were characterized with an impedance analyzer from 20 Hz to 1 MHz and analyzed as a resistor R in parallel with a capacitor C. The calculated noise voltage spectral density S_vTh of the sample is in agreement with the experimentally observed noise for unbiased samples.

The noise measurements on biased samples show 1/f noise. We checked that contact noise does not contribute to our measurements. The 1/f noise was compared for the films deposited on the silicon substrate and on the sapphire.

Finally a model is proposed to explain the observed trends in the temperature resistance coefficient (TRC < 0 for all samples) and the relative 1/f noise at 300 K and 80 K with the composition x = N/Nb of the layers.

The correlation integral is the main tool for the analysis of the chaotic time series. Its linear part (i.e. the scaling region) gives the attractor's dimension. In this work we propose a model that explains all the possible neighborhood ranges of the correlation integral. Empirical examples are given on artificial data.

Motivated by biological neural networks and distributed sensing networks, we study how pooling networks – or quantizers – with random thresholds can be used in detection tasks. We provide a brief overview of the use of deterministic quantizers in detection by presenting how quantizers can be optimally designed for detection purposes. We study the behavior of these networks when they are used in a problem for which they are not optimal (mismatching). We show that adding random fluctuations to the thresholds of the networks can then enhance the performance of the quantizers, thus helping in the recovery of "a kind of" optimality. We also show that (for a small number of thresholds) it suffices to use random uniform quantizers, for which we provide a study of the behavior as a function of several parameters (size, fluctuation nature, observation noise nature). The conclusion to these studies are the robustness of the uniform quantizer used as a detector with respect to fluctuations added on its thresholds.

The present paper investigates the influence of environmental noise on a fairly realistic three-species food chain model based on the Leslie-Gower scheme. The self- growth parameter for the prey species is assumed to be perturbed by white noise characterized by a Gaussian distribution with mean zero and unit spectral density. Using tools borrowed from the nonlinear dynamical system theory, we study the dynamical behavior of the model system. The behavior of the stochastic system (perturbed one) is studied and the fluctuations in the populations are measured both analytically (for the linearized system) and numerically by computer simulation. Varying one of the control parameters in its range, while keeping all the others constant, we monitor the changes in the dynamical behavior of the model system, thereby fixing the regimes in which the system exhibits chaotic dynamics. Our study suggests that the trophic level (top, middle or bottom) at which a population is positioned, the amplitude of environmental noise and the population's susceptibility to environmental noise play key roles in how noise affects the population dynamics.

If data is binary, it is probable that the combination of a signal of interest plus a noise has been simplified by a thresholding mechanism, as in, e.g., a neuron firing mechanism. For identifying optimal signal or coding range of binary data, Fisher information is an attractive measure. A general formula allows the signal level or signal range producing the most information-rich data to be identified if the noise distribution is known. In this paper we study the information content of binary data resulting from threshold exceedance of a signal plus an arbitrary type of noise. For a specified parametric family of distributions a fixed proportion of exceedances is optimal for any combination of signal, threshold, and noise amplitude. If the ratio of noise to signal level is constant, Fisher information is unimodal for many noise distributions. The results extend to the case of a random signal and to inter-exceedance-interval data. The family of gamma noise distributions is used for illustration.

We report on the degradation of low-frequency excess noise in recessed gate AlGaN/GaN HEMTs due to hot-electron stressing. The I-V characteristics and the low-frequency noise power spectral densities, S_V(f), of the open circuit voltage fluctuations across the drain source terminal were characterized with the stress time. Based on these results, we observed that the overall low-frequency noise degradation process can be identified to occur in two distinct phases. In the first phase, devices initially show fluctuations in the noise properties around a relatively constant average value. Detailed characterizations of the gate-source bias, V_GS, dependence of S_V(f) at cryogenic temperatures indicate signature-patterns in the variations of S_V(f) as a function of V_GS. This is shown to arise from the modulation of the percolation paths of the carriers in the two-dimensional electron gas (2DEG). The onset of the second phase of degradation arises from the irreversible generation of interface states at the AlGaN/GaN hetero-interface.

We review existing methods for generating long streams of 1/f^α noise (0 < α ≤ 2) focusing on the digital filtering of white noise. We detail the formalism to conceive an efficient 1/f^α random number generator (white outside some bounds) in order to generate very long streams of noise without an exhaustive computer memory load. For α = 2 it is shown why the process is equivalent to a random-walk and can be obtained simply by a first order filtering of white noise. As soon as α<2 the problem becomes non linear and we show why the exact digital filtering method becomes inefficient. Instead, we work out the formalism of using several 1/f² filters spaced logarithmically, to approximate the spectrum at the percent level. Finally, from work on logistic maps, we give hints on how to design generators with α > 2. The corresponding software is available from .