Wideband Random Noise Research Articles

Weak Fault Feature Extraction of Rotating Machinery Based on Double-Window Spectrum Fusion Enhancement

Early weak fault feature extraction is difficult due to the interference by wideband random noise and other vibration sources in the same frequency band. In this paper, a weak fault feature extraction method of rotating machinery based on the double-window spectrum fusion enhancement (DWSFE) is proposed to address this issue. This method is established through combining fast Kurtogram algorithm and spectrum fusion enhancement. In this scheme, the fast Kurtogram algorithm is utilized to suppress the interference of weak fault feature extraction by noise signals outside the optimal resonant band and to achieve the signal envelope demodulation in the passband of the optimal resonant region. Subsequently, discrete short-time Fourier transforms (DSTFTs) with long- and short-time windows are constructed to obtain the time–frequency spectra with different time–frequency resolutions, respectively. By integrating the advantages of the time–frequency resolutions of these spectra, the spectrum correlation analysis denoising and the spectrum amplitude association enhancement are proposed to suppress the random interference in the passband and to highlight the signal characteristic frequency. Furthermore, the weak fault features are separated from other interference of vibration sources in the same frequency band, and finally, the weak fault feature extraction of rotating machinery can be achieved. The simulation and application results show that this method can enhance weak fault features while denoising and achieve the weak fault feature extraction of rotating machinery under strong noise.

Корреляционная оценка при передаче информации относительным методом на основе широкополосных шумовых сигналов.

Корреляционная оценка при передаче информации относительным методом на основе широкополосных шумовых сигналов.

Noise interference with echo delay discrimination in bat biosonar.

Echolocating big brown bats (Eptesicus fuscus) were trained in a two-choice task to discriminate differences in the delay of electronic echoes at 1.7 ms delay (30 cm simulated range). Difference thresholds (∼45 μs) were comparable to previously published results. At selected above-threshold differences (116 and 232 μs delay), performance was measured in the presence of wideband random noise at increasing amplitudes in 10-dB steps to determine the noise level that prevented discrimination. Performance eventually failed, but the bats increased the amplitude and duration of their broadcasts to compensate for increasing noise, which allowed performance to persist at noise levels about 25 dB higher than without compensation. In the 232-μs delay discrimination condition, echo signal-to-noise ratio (2E/N0) was 8-10 dB at the noise level that depressed performance to chance. Predicted echo-delay accuracy using big brown bat signals follows the Cramér-Rao bound for signal-to-noise ratios above 15 dB, but worsens below 15 dB due to side-peak ambiguity. At 2E/N0 = 7-10 dB, predicted Cramér-Rao delay accuracy would be about 1 μs; considering side-peak ambiguity it would be about 200-300 μs. The bats' 232 μs performance reflects the intrusion of side-peak ambiguity into delay accuracy at low signal-to-noise ratios.

Local Similarity in Nonlinear Random Vibration

A response analysis procedure is developed for oscillators with highly nonlinear stiffness and light nonlinear damping excited by non-white wide-band random noise based on local similarity between the random response and the deterministic response at the same energy level of the corresponding undamped oscillator. The analysis consists of three parts: introduction of modified phase plane variables, derivation of an approximate general form of the probability density of the response energy. for non-white excitation, and derivation of the spectral density function of the response from the conditional covariance function for a given energy level. The use of modified phase plane variables leads to a completely symmetric formulation and reformulates the stiffness nonlinearity as a nonlinear variation of the instantaneous angular frequency, and thereby a local rescaling of time. The probability density is obtained by averaging the full Fokker-Plank-Kolmogorov equation using local similarity, thus avoiding some theoretical problems associated with the traditional averaging of the stochastic differential equations. The use of local similarity with the exact undamped solution in the derivation of the conditional spectral density leads to a spectral density estimate, that contains the higher harmonic components explicitly. Comparisons of theoretical predictions with digital simulation estimates of both the probability and spectral densities for the Duffing oscillator demonstrate the accuracy of the theory.

Split structure for adaptive line enhancer

A new adaptive line enhancer (ALE) is suggested to speed up the convergence rate for the retrieval of multiple sinusoids corrupted by a wide-band random noise. The proposed ALE structure comprises two transversal linear phase filters connected in parallel, one being antisymmetric and the other symmetric. The two filters are adapted separately using a least-mean-squares algorithm. The convergence characteristics and computational requirements of the new ALE are examined in detail. Simulation results have shown that the split structure ALE can significantly improve the rate of adaptation while still maintaining system simplicity for real-time hardware implementation.

Photopyroelectric thin-film instrumentation and impulse-response detection. Part III: Performance and signal recovery techniques

Frequency-modulation time-delay spectrometry (FM-TDS) was compared with stochastic excitation in photopyroelectric measurements on thin solid samples. These methods involve broadband signal excitation with simultaneous detection of all frequency components within the response bandwidth of the photothermal system. Fast signal recovery was achieved by FFT methods. Both FM-TDS and wideband random-noise excitation were found to yield high-quality, band-limited impulse-response information. Random noise measurements were found to be much less susceptible to distortions and nonlinearities in the excitation wave train. FM-TDS showed superior coherence and signal-to-noise ratio (SNR) to random methods. The results of this work demonstrate a photopyroelectric effect spectrometer capable of yielding fast, time-resolved information equivalent to the response of a pulsed laser, with low peak power. This important feature shows excellent potential for nondestructive thickness, thermal diffusivity, and thermal conductivity measurements on materials which are susceptible to optical damage.

Traffic-alarm sound monitor for aurally handicapped drivers.

There are several countries in the world in which people with severe hearing loss are not eligible for a car driver’s licence. As a technical approach to solve this problem, an electronic device has been developed which detects traffic-alarm sounds, i.e. horns of cars, sirens of emergency vehicles, and alarm signals of railway crossings, and then displays them as a light signal to the driver. The basic operating principle of the device is that those traffic-alarm sounds have sharp line spectra in the frequency domain whereas ambient traffic noise is wide-band random noise. The real time detection of the line spectra, masked by random noise, is realised by use of a phase-locked loop and a simplified lock-in amplifier. The results of simulation experiments and road tests demonstrate that the performance of the device is satisfactory except in the case of the detection of the alarm signal of a railway crossing.

Quality rating of television pictures: comparison of results obtained by laboratories in different countries

The participation of five European countries in preliminary experiments with methods of television subjective testing, intended for application to a 2 Mbit/s codec, has provided the first opportunity for comparing results obtained by different organisations using a single-stimulus full-range quality-rating method under fairly well standardised conditions. The range in standards of subjective judgment was about 2 dB, as expressed in terms of the equivalent level of wideband random noise. This small range, coupled with simplicity, encourages the continued use of quality grading by the single-stimulus method for television subjective studies involving a substantial range of subjective magnitudes.

Atmospheric absorption adjustments for bands of noise

A new procedure has been developed for determining the absorption experienced by pure‐tone sound waves propagating through the atmosphere. The procedure, which accounts for classical absorption processes (viscosity and heat conduction) as well as molecular energy transfer mechanisms, has a firm theoretical basis, has been extensively verified by laboratory experiment, and has been issued recently as an American National Standard ANS 1.26‐1978. To apply the procedure of ANS 1.26 to determination of atmospheric absorption losses experienced by wideband random noise required development of a method that took into account the slope of the power spectrum of the sound pressure at the microphone. An important application of the method would be to adjust a measured 13‐octave‐band sound‐pressure‐level spectrum for differences in atmospheric absorption occurring under test and reference meterological conditions. Development and verification of the method are described for application to measurements of aircraft flyover noise. [Work supported by the Federal Aviation Administration.]

Calculation of wideband random-noise powers in television systems

Expressions are derived for the loss experienced by wideband random noise with various power-density spectra in networks associated with television transmission systems. Values are tabulated for television system I with the total noise power modified by each of two weighting networks used in noise measuring circuits and/or the de-emphasis network used to complement pre-emphasis with f.m. radio transmission.

Some aspects of sweep random vibration

Following a brief description of the basic reasons for sine-wave and wide band random vibration testing some general arguments in favour of a new vibration test technique are presented. It is proposed that the new test may, in many cases, give a better simulation of the expected vibration environments than any of the two previously used tests. The new test, normally called sweep random vibration, consists in sweeping one or more bands of random noise through the frequency range of interest; the vibration level may be changed during the sweep according to any predetermined level vs. frequency test programme and the advantages of simple servo-controlled level regulation are utilized. Methods for reducing the test time as well as possible setting-up schedules are discussed. Due to the reduction in the requirements to power handling capacity of the vibration machine and power amplifier when the new test is specified, greater test specimens can be conveniently tested on existing laboratory test equipment than if wide band random noise is used.

Automatic Equalization of Resonances in Sonic Fatigue Chambers

As constructed, all acoustical chambers and most acoustic transducers exhibit sharp resonant peaks, resulting in a concentration of energy over a few narrow frequency bands. Furthermore, these resonant frequencies vary for different specimens placed in a chamber. In order to achieve the desired “flatness” with respect to frequency it is necessary to create electronically the inverse of this frequency characteristic at the input to the system. Early types of equalizing circuits required laborious manual adjustment of many interacting controls to achieve equalization for each new specimen. The instrument described in this report automatically compensates the chamber for both sine and random wave testing. The system consists of 27 separate closed servo loops, each one controlling a 13 octave segment of the total frequency band. The resulting amplitude-controlled segments are then summed and applied to the acoustic driver. The problem of phase cancelation between adjacent 13 octave bandpass filters is overcome through the use of special linear-phase bandpass filters and a toroidol phase compensating transformer. By using a unique solid-state circuit to compensate each frequency segment, it is possible to equalize all resonance occurring in a 200-cu-ft acoustic chamber within 10 sec after the application of a wide-band random noise input.

Effect of Stimulus Duration on Lateralization Threshold

Measurements on experienced listeners of interaural time difference thresholds for wide-band random noise indicated that the threshold varies systematically with duration of the stimulus. In order to determine the point at which increase in duration no longer decreases interaural time difference threshold, stimulus (noise burst) duration was varied between 0.01 and 1.9 sec. Fortunately, the use of wide-band noise made it practical to employ instantaneous (but not simultaneous) rise times, and thus to specify duration precisely. A given interaural time difference was maintained throughout any particular burst, starting time included. The effects of simultaneous (and even reversed) onsets, of band-width and intensity variation are reserved for further study. Throughout the study, all stimuli were presented at a level of 65 db SPL to each phone. The duration versus interaural time difference threshold function reached asymptote at approximately 0.7 sec, indicating that the binaural system which effects the correlation necessary for a lateralization judgment may integrate information over this long a period for the kind of stimulus used.