On quantization of noisy signals

Quantized noise distributions derived from continuous signals with additive noise are studied. Two noise sources are considered, quantized image noise derived from the continuous input noise source and noise due to quantization roundoff error. These are treated as statistically independent sources. An analytic solution for the quantized noise probability density is obtained. The analytic solution is estimated by two expressions valid for normally distributed noise over different ranges of variance. The estimates have excellent agreement in the region of overlapping validity. Quantized noise variance is related to the continuous noise variance from normally distributed noise using these expressions. A table and plots of useful values are included. These results are helpful in choosing a quantization interval for a particular application. They can also be used to determine quantizer output noise level and signal-to-noise ratio in digital applications     

Compact Noise Models for MOSFETs

A physical understanding of both intrinsic and extrinsic noise mechanisms in a MOSFET is developed. Intrinsic noise mechanisms fundamental to device operation include channel thermal noise, induced gate noise, and induced substrate noise. While the effect of channel thermal noise is observable at zero drain-to-source voltage, the induced gate and substrate noise do not manifest themselves under these conditions. However, the attendant fluctuations in the channel charge are observable by the passage of electric current through the device. Extrinsic noise mechanisms manifested due to structural evolution of the MOSFET include the distributed gate resistance noise, distributed substrate resistance noise, bulk charge effects, substrate current supershot noise, gate current noise, excess channel noise, and$hbox1/f$noise. Where available, compact noise models covering these noise mechanisms are explained. Also, where possible, methods of suppression of these mechanisms are highlighted. A survey of current public domain MOS models is presented, and a lack of comprehensive coverage of noise models is noted. Open areas of MOSFET noise research in the sub-hundred-nanometer regime are also highlighted. With suitable adaptation, noise concepts elucidated in the context of MOS transistors have a much wider applicability to the operation of HEMTs, JFETs, MESFETs, and other field-effect devices.     

The active device characterization and modelling problem in low phase noise microwave oscillator design

The conversion of low frequency noise into phase noise in a high frequency oscillator depends strongly on the active device type and operating conditions. Experimental set-ups are presented to achieve the choice of a transistor to be used in low noise oscillators and to study the correlation between phase noise and low frequency noise. Modelling guidelines are then described to help in the understanding of the noise conversion process. Simple rules of noise reduction are derived from simplified small signal models but the emphasis is put on more complete modelling approaches. These accurate models are affected by yet unsolved (or partially unsolved) problems such as : the influence of the large signal behaviour on the noise sources or the physically unacceptable cancellation of the phase noise in special biasing conditions.     

A study of phase noise in CMOS oscillators

This paper presents a study of phase noise in two inductorless CMOS oscillators. First-order analysis of a linear oscillatory system leads to a noise shaping function and a new definition of Q. A linear model of CMOS ring oscillators is used to calculate their phase noise, and three phase noise phenomena, namely, additive noise, high-frequency multiplicative noise, and low-frequency multiplicative noise, are identified and formulated. Based on the same concepts, a CMOS relaxation oscillator is also analyzed. Issues and techniques related to simulation of noise in the time domain are described, and two prototypes fabricated in a 0.5-μm CMOS technology are used to investigate the accuracy of the theoretical predictions. Compared with the measured results, the calculated phase noise values of a 2-GHz ring oscillator and a 900-MHz relaxation oscillator at 5 MHz offset have an error of approximately 4 dB     

基于细节保留的椒盐噪声自适应滤波算法

针对灰度图像中椒盐噪声的特点,提出了一种更加精确的噪声检测方法:该方法利用滤波窗口内像素点灰度值的不同,将受椒盐噪声污染的图像中像素点划分为噪声点,疑似噪声点和信号点.通过设定阈值,并参考相邻像素点的相关性来进一步区分疑似噪声点,最终建立噪声标记矩阵.对于被标记的噪声点,采用自适应滤波算法,保留更多的图像细节.仿真结果表明,该算法在除去噪声点的同时,对于边缘细节也有非常好的保护作用.     

Matrix Representations of Noise Figures and Noise Figure Charts in Terms of Power Wave Variables

The noise power, excess exchangeable noise figure, and excess transducer noise figure of a linear active two-port are represented by the power wave scattering transfer matrix. It follows that these noise figures are expressed as ratios of two Hermitian forms, and that their stationary values and the reflection coefficients at the input which yields them are obtained from the eigenvalues and the corresponding eigenvectors of each characteristic matrix. The excess transducer noise figure is considered here only for the case of the load being a resistor. It is proved that the transducer noise figure is equal to the exchangeable noise figure, which includes the contribution of equivalent noise at the input transformed through a two-port from the noise originating in the load. In order to evaluate graphically the change of noise figure of a two-port from its optimum value because of the reflection coefficient at the input, two kinds of noise figure charts are developed: one shows the contours of the constant noise figure normalized by the optimum value, and the other shows the contours of the constant difference between the prescribed noise figure and the optimum value. Their application to a practical problem is demonstrated. The noise figure of a linear active three-port with a passive termination /spl Zeta//sub 3/' at port 3 is expressed by a ratio of two Hermitian forms, and its stationary value and the reflection coefficient /spl Gamma/ /sub 3/ of /spl Zeta//sub 3/' which yields it are obtained. Two noise figure charts are presented by means of which the noise figure of a three-port for a specified /spl Gamma/ /sub 3/ can be evaluated.     

Channel noise modeling of deep submicron MOSFETs

This brief presents a new channel noise model using the channel length modulation (CLM) effect to calculate the channel noise of deep submicron MOSFETs. Based on the new channel noise model, the simulated noise spectral densities of the devices fabricated in a 0.18 /spl mu/m CMOS process as a function of channel length and bias condition are compared to the channel noise directly extracted from RF noise measurements. In addition, the hot electron effect and the noise contributed from the velocity saturation region are discussed.     

A Systematical Approach for Noise in CMOS LNA

A systematic approach is used to analyze the noise in CMOS low noise amplifier(LNA),including channel noise and induced gate noise in MOS devices.A new analytical formula for noise figure is proposed.Based on this formula,the impacts of distributed gate resistance and intrinsic channel resistance on noise performance are discussed.Two kinds of noise optimization approaches are performed and applied to the design of a 5.2GHz CMOS LNA.     

Bistable noise in operational amplifiers

Bistable noise in operational amplifiers has been investigated. Equivalent circuits are used. It is shown that by changing the source resistance of an operational amplifier the locations of bistable noise sources can be determinated. Both burst noise and microplasma noise observed from an operational amplifier. The input transistors of an operational amplifier are the most important bistable noise sources. Burst noise due to elements other than the input transistors is shown to be important. Surface effects on burst noise have been discussed. Because of the surface effect, n-p-n transistors are expected to show more burst noise than p-n-p transistors. Microplasma noise in an operational amplifier is attributed to a local surface breakdown.     

Optimal low-frequency noise criteria used as a reliability test for BJTs and experimental results

Two basic problems in the application of noise criteria to reliability testing are considered. First, the need to look at all noise mechanisms rather than noise at a specific frequency is emphasized, and noise criteria using both 1/f noise and g−r noise are established. Second, the physical mechanisms which explain the low noise level of some failed devices are discussed, and the optimal noise threshold levels are found for minimum error during reliability testing.Reliability testing results for 1000 BJTs (3DG6) show that the failure rate of devices which initially exhibit high noise is about 2–3 times higher than for devices which initially have low noise. This means that the life expectancy of BJTs with a low noise level (using optimal noise criteria) is increased by a factor of 1.5–2.5; thus, noise measurements could be a useful reliability screen classification method for BJTs.     

Modal noise in multimode fibers under restricted launch conditions

Modal noise is a non-Gaussian noise source associated with the interference of fiber modes at a detector. Most derivations of the statistics of modal noise assume a limiting case of overfilled launch conditions where there are many modes propagating in the fiber. However, if the transmitted power is contained in a few modes, commonly referred to as restricted-mode launch conditions, the general assumptions for modal noise theory are not valid. Here, we present experimental measurements of the noise distributions for both full- and restricted-mode launch cases and show that the noise statistics for a restricted launch are significantly different than for a full-mode launch. In particular, the variance of the noise distribution is larger than in the over-filled launch case leading to a degradation in the signal-to-noise ratio (SNR). Modifications of existing theory are then used to develop a numerical model for the noise statistics of a restricted-mode launch and good agreement between this model and experimental data is achieved     

CMOS低噪声放大器中噪声的系统研究方法

采用系统研究方法来分析包括MOS器件的沟道噪声和感应栅噪声在内的CMOS低噪声放大器中的噪声,并提出了一个新的噪声系数解析式.基于此解析式,讨论了分布栅电阻和内部沟道电阻对噪声性能的影响.对噪声性能进行了两种不同的优化,并应用于5.2GHz CMOS低噪声放大器的设计.     

凝视热成像系统三维噪声分析法

三维噪声技术是分析凝视热成像系统噪声模型的一种新方法,与传统的噪声分析方法比,这种方法更能体现凝视热成像系统的噪声特征.文中介绍了三维噪声技术的分析方法和理论框架,分析了传统噪声在三维噪声中的体现,并简单论述了三维噪声的有关应用.     

Noise cross PSD estimation using phase information in diffuse noise field

Among various speech enhancement methods, two-microphone noise reduction systems are utilized for their low cost implementation and acceptable performance. Coherence-based methods are well known as efficient two-microphone noise reduction techniques. These techniques, however, do not work well when the received noise signals are correlated. Coherence-based methods can be improved when the cross power spectral density (CPSD) of input noises is available. In this paper, we propose a new method for estimating noise CPSD based on the assumption of a diffuse noise field. With this assumption, we estimate the noise CPSD using phase information. Then, the estimated noise CPSD is used to calculate a coherence-based gain filter which is then employed to enhance noisy signals. We compare the proposed phase-based noise CPSD estimation with a noise CPSD estimation technique based on a voice activity detector (VAD), both of which are herein separately employed in a two-microphone speech enhancement configuration. The comparison shows that the two-microphone speech enhancement scheme utilizing the proposed noise CPSD estimation technique outperforms the enhancement system using the VAD-based noise CPSD estimation.     

Design and noise analysis of a sigma-delta capacitive micromachined accelerometer

A single-loop fourth-order sigma-delta (∑△) interface circuit for a closed-loop micromachined accelerometer is presented.Two additional electronic integrators are cascaded with the micromachined sensing element to form a fourth-order loop filter.The three main noise sources affecting the overall system resolution of a ∑Δ accelerometer,mechanical noise,electronic noise and quantization noise,are analyzed in detail.Accurate mathematical formulas for electronic and quantization noise are established.The ASIC is fabricated in a 0.5 μm two-metal two-poly n-well CMOS process.The test results indicate that the mechanical noise and electronic noise are 1 μg(Hz) and 8 μV/(Hz)respectively,and the theoretical models of electronic and quantization noise agree well with the test and simulation results.     

A Note on Noise Temperature

The effective noise temperature of the output impedance of a Iossy passive network at an arbitrary noise temperature connected to one or more resistive loads at arbitrary noise temperature lies between the highest and the lowest of these noise temperatures, as determined by the losses between the output terminals and the loads. The determination of the effective noise temperature of a gas-discharge noise generator over a wide frequency range is simplified by the substitution of a loss measurement for the more difficult noise temperature measurement. For minimum-noise radar applications care must be used in considering the excess noise of crystal mixers and gas-discharge duplexers. The influeuce of galactic radiation on a receiving system is such that there is an optimum frequency in the region of 200 to 600 mc for minimum "operating noise figure." Typical examples of radio-astronomy measurements are amenable to analysis of the type given. Finally, several corrections to measured noise figure are analyzed.     

Modeling of noise parameters of MESFETs and MODFETs and theirfrequency and temperature dependence

A simple noise model of a microwave MESFET (MODFET, HEMT, etc.) is described and verified at room and cryogenic temperatures. Closed-form expressions for the minimum noise temperature, the optimum generator impedance, the noise conductance, and the generator-impedance-minimizing noise measure are given in terms of the frequency, the elements of a FET equivalent circuit, and the equivalent temperatures of intrinsic gate resistance and drain conductance to be determined from noise measurements. These equivalent temperatures are demonstrated in the case of a Fujitsu FHR01FH MODFET to be independent of frequency in the frequency range in which 1/f noise is negligible. Thus, the model allows prediction of noise parameters for a broad frequency range from a single frequency noise parameter measurement. The relationships between this approach and other relevant studies are established     

Measurement and model for correlating phase and baseband 1/f noisein an FET

Phase noise in solid state oscillators arises in part from residual phase noise in the active device that provides gain. The device's residual phase noise is usually attributed to baseband noise on the bias that upconverts via a mixing process to generate phase noise around the carrier signal. This conclusion is consistent with the observation that baseband and carrier noise sources both have 1/f spectral properties. However, similar spectral properties are not sufficient to prove that the baseband and carrier noise are fully-correlated. Experimental results presented here for a heterojunction FET show less than 40% correlation. To account for these observations a new 1/f noise model is presented. This model is used to accurately predict the phase noise when a bias feedback circuit is added to stabilize the baseband fluctuations     

Noise in Single-Frequlency Oscillators and Amplifiers

A generalization of previous oscillator noise analyses has been developed to permit reliable noise characterization of active nonlinear devices. Effects due to sideband correlation in the equivalent noise source are included. A rotating wave approximation (RWA) developed by Lax is used in obtaining the amplitude and phase noise spectra. Conditions are given for phase stabilization of free-running oscillators and for minimum phase noise in phase-Iocked oscillators and amplifiers. Stability criteria, discussion of spurious sidetones, and effects of a noisy synchronizing signal are given. The noise measure is used to obtain alternative expressions for the noise spectra and the carrier-to-noise ratios of locked oscillators and amplifiers. It is shown that the noise power gain of AM fluctuations is usually much lower than the corresponding gain for FM noise. The theory should be useful in optimizing the noise performance of nonlinear RF generators, such as IMPATT, BARITT, and Gunn diode oscillators.     

Noise analysis of an amplified fiber-optic recirculating-ring delayline

We present the first theoretical and experimental noise analysis of a fiber-optic recirculating-ring delay line (RDL) including a doped fiber amplifier to compensate for the roundtrip loss. Both thermal-like sources and laser sources are considered. The output source induced noise (signal-signal beat noise), signal-spontaneous (s-sp) beat noise, and spontaneous-spontaneous (sp-sp) beat noise spectra for a thermal-like source are calculated from the autocorrelation function of the output detector current. It is shown that all three electrical beat noise spectra can be expressed as correlations of the output optical signal and ASE spectra. The source-induced noise will normally be the dominating noise source, but in some applications, the other noise terms also will be of importance. We use our theory to define the maximum number of recirculations in an amplified RDL with a pulsed source, where the fundamental noise floor is determined by the sp-sp beat noise