No description available.

No description available.

Thesis. 1975. M.S.--Massachusetts Institute of Technology. Dept. of Aeronautics and Astronautics. / Includes bibliographical references. / M.S.

No description available.

The work reported here is an experimental and theoretical investigation of high-frequency electrical noise generated in extrinsic single-crystal n-type germanium at high electric fields. The electric field was pulsed so that the lattice temperature remained near 77°K. During the pulse, the electrons quickly reach a non-equilibrium steady-state due to their gaining energy from the electric field and brought to a steady-state by means of collisions with the lattice vibrations. Previous work has been concerned with noise measurements made at right angles to the electric field direction, where anisotropic behaviour was observed. The present measurements, made in the direction of the electric field, also show a high degree of anisotropy. The electrical noise generated is described by the noise temperature, Tn, obtained by adapting the Nyquist formula to the non-equilibrium case. Measurements of Tn, performed at frequencies of 70Mc/s and 30Mc/s, indicated a uniform noise spectrum in this frequency range for all the samples used. The anisotropy of Tn suggested that Tn was explainable on the basis of the many-valley model of the conduction band of germanium, established by previous experimental investigations of the high-field mobility anisotropy. A feature of the many-valley model is that electrons in different valleys of the conduction band, will in general, exhibit different transport behaviour and as a result, transitions between these valleys will result in a noise phenomenon described as "intervalley noise". However, for measurements of Tn in the <100> direction, the "intervalley noise" will vanish, allowing a direct measure of the electron "heating" due to the electric field, the “hot electron noise”. In the <111> an <110> directions, both intervalley and hot electron noise are expected. Both contributions to Tn are evaluated by means of Barrie's extension to the case of many-valley germanium of Stratton's high-field transport theory. / Science, Faculty of / Physics and Astronomy, Department of / Graduate

In the past, spectral analysis of Arctic Sea ambient noise has been carried out by octave band pass filtering and linear rectification, followed by analogue Integration. To relate this integral to the power spectral density of the noise, the amplitude distribution of the noise has been assumed to be Gaussian, giving rise to certain errors. A system is proposed which consists of a stage of variable gain followed by a polarity coincidence statistical wattmeter and measures the power spectral density of ambient noise after band pass filtering. The wattmeter will handle an input signal dynamic range of at least 20 dB and does so regardless of the statistical nature of the noise. This dynamic range is extended dynamically by controlling the gain of the driving stage. The gain level is automatically adjusted during a one minute "adaptive" time interval so that the noise delivered to the wattmeter is over the region of optimal system operation. Measurement of the power spectral density of the ambient noise is then made in the subsequent four minute interval. A prototype wattmeter has been constructed and tested. The gain level is determined by requiring that the noise not exceed fixed levels more than a certain percentage of the time. This automatic adjustment is carried out during a one minute adaptive time interval, and a relatively accurate measure of the mean square value of the noise is determined during the four minutes that follow. For purposes of testing the prototype, d.c. inputs and sinusoidal inputs of wide frequency and amplitude ranges were used. The actual root mean square value of the inputs was measured with a thermal milliammeter and a precision voltage divider. The results of these tests show the region of operation where the input-output relationship of the wattmeter is linear. From these results, suggestions are made.as to how the proposed system could be modified to replace the analogue system used for Arctic Sea ambient noise spectral analysis. / Applied Science, Faculty of / Electrical and Computer Engineering, Department of / Graduate

This thesis presents distinctly different methods of accurately predicting phase noise and absolute jitter in ring oscillators. The phase noise prediction methods are the commercially available SpectreRF and isf_tool, a simulator developed in this work from the Hajimiri and Lee theory of phase noise. Absolute jitter due to deterministic supply and substrate noise is predicted by Spectre time domain simulations and equations developed that can predict the absolute jitter due to a sinusoidal noise source at any frequency. These jitter prediction methods show that ring oscillator circuits respond differently to deterministic noise that is injected symmetrically versus noise that is injected asymmetrically, and a new jitter metric, peak jitter, is developed in this work to characterize absolute jitter caused by deterministic noise sources. These prediction methods are validated with measurements from two test chips with a combined 18 oscillators and 5 distinct architectures, and both are fabricated in the TSMC 0.35μm process. Each prediction method is shown to be consistent with over 2500 phase noise measurements taken from 10 oscillators and 5 architectures and over 1200 absolute jitter measurements due to sinusoidal supply and substrate noise taken from 11 oscillators and 3 architectures. / Graduation date: 2002

The purpose of this study was to indicate norms for most comfortable loudness levels (MCLs) and to investigate the relationships among MCL and related measures. In addition MCL values for males and MCL values for females were separated to see if any differences existed between sexes.A pure tone average (PTA) and a spondee threshold (ST) were obtained for each of 54 subjects (23 males, 31 females) to assure normal hearing. The following measures were then obtained on all subjects: 1.) The MCL was obtained using a bracketing method. 2.) A Bekesy audiometer was used to determine MCL using both a 2.5 dB/second and a 5 dB/second attenuation rate. 3.) The lower limit of the most comfortable loudness range (LLMCR) was determined using a nonBekesy procedure. 4.) The upper limit of the most comfortable loudness range (MCR). Again, both 2.5 dB/second and 5 dB/second attenuation rates were used. The Bekesy MCR tracings also provided the ULMCR and the LLMCR values for Bekesy obtained MCRs. Four analyses of variance (ANOVA) were computed for various MCL and MCR measures to investigate the relationships among these measures. It was concluded that no significant differences occurred between males and females in all MCL and related measures.Averages were presented for each individual MCL procedure which ranged from 51.29 dB to 56.77 dB HL. Average MCRs (actual ranges) were 30.54 to 42.78 dB. Some significant differences were found among specific procedures which are discussed in the text, however, these differences may not be significant from a clinical standpoint. In general, non-Bekesy procedures were recommended for use clinically because they produced the least variability in subjects and they require the least amount of time in administration.

A dissertation submitted in fulfllment for the degree of Masters of Science in the Faculty of Science School of Computer Science and Applied Mathematics October 2017. / This work explores a new method of applying partial di erential equations to audio signal processing, particularly that of noise removal. Two methods are explored and compared to the method of noise removal used in the free software Audacity(R). The rst of these methods uses a non-linear variation of the di usion equation in two dimensions, coupled with a non-linear sink/source term, in order to lter the imaginary and real components of an array of overlapping windows of the signal's Fourier transform. The second model is that of a non-linear di usion function applied to the magnitude of the Fourier transform in order to estimate the noise power spectrum to be used in a spectral subtraction noise removal technique. The technique in this work features nite di erence methods to approximate the solutions of each of the models. / LG2018

No description available.