基于音频信号的量化噪声教学实验

量化噪声的产生及其特性，是数字信号处理课程中一个贴近工程实践的重要知识点。有别于传统理论讲授模式，本文设计了一个“量化噪声响起来、看得见”的教学实验，分别通过截尾和舍入两种量化方式对音频信号进行处理，不仅可以让学生直观感受量化噪声的存在与特性，还可以更深刻理解量化方式、量化位数等因素对量化噪声特性的影响。     

基于均匀量化器的随机量化方法

从易于硬件实现的角度，提出了一种基于均匀量化器的随机量化方法。证明了该方法在所加噪声满足一定条件时，其量化信号的数学期望等于相应的量化输入值。分析了所加噪声对量化动态范围的影响。利用计算机模拟，对量化输出平均值的分布规律及其与反复量化次数和所加噪声取值域的关系进行了分析研究，为以后设计量化器提供了依据。     

基于CAN总线的标量控制系统中动态量化算法研究

在基于CAN总线的标量控制系统中引入动态量化算法,可以有效降低总线负载,但是也会引入量化误差。在受有界噪声影响下的标量控制系统中,其量化误差不能收敛到0。文章通过分析得知量化噪声可以降低有界噪声的影响。基于此,首先提出了噪声标志位算法,随后分析了单比特噪声标志位算法的性能,最后给出了最优比特分配方案来获得更低的量化误差。     

低位量化对数字化多频连续波雷达测距性能的影响

对基于FFT数字比相技术的多频连续波雷达，量化特别是低位量化必然对雷达的测距精度具有一定的影响。首先给出了极小化量化处理的实现方法，然后比较了极小化量化与均匀量化在同等条件下的量化噪声的方差及信噪比损失，推导出考虑量化噪声时多频连续波雷达测距误差均方根的计算公式，最后通过仿真结果验证了低位量化噪声对测距精度的影响，尤其是1位或2位量化对测距精度的影响。     

Sigma-delta电容式微机械加速度机设计与噪声分析

设计了一种单环4阶∑△加速度计的接口专用集成电路（ASIC）,采用敏感结构和二阶开关电容积分器组成4阶积分结构。详细分析了影响∑△微机械加速度计系统精度的三种噪声：机械噪声、电路噪声和量化噪声,建立了电路噪声和量化噪声的精确的数学公式。该电路采用0.5um 两层金属两层多晶n阱CMOS工艺流片,测试结果表明机械噪声和电路噪声分别为1ug/rtHz和8uV/rtHz,电路噪声和量化噪声的理论模型预测值与测试和仿真结果相一致。     

数字音频信号信噪比的提高

影响数字声频质量的主要原因之一是量化噪声。本文讨论了采用过取样和噪声成形等方法，来降低数字音频信号中的量化噪声，这些方法还可使模拟低通滤波器大大简化，此外，采用低比特数再量化，可以降低模数变换器及数模变换器的价格。     

重复观测提高量化精度方法的研究

模数转换的精度在很大程度上取决于模数转换器的位数。通过在量化前使被量化信号叠加上适当分布的噪声并对量化值作多次取样平均，有可能提高量化精度。本文通过严格的数学分析，给出了估计无偏的条件以及为满足给定量化精度所需的最少积累次数。最后对采用正态分布和均匀分布噪声作为叠加噪声的情形进行了具体的分析和比较。     

数字化频率综合器的相位噪声分析与估算

利用随机过程理论计算了数字化频率综合器中各个量化噪声，并在假设各等效量化噪声相互独立前提下得出了输出相位噪声值，同时给出在不同采样率和不同数字电路精度情况下输出相位噪声变化，并得出了一般性结论。     

一种基于LMS自适应滤波的CVSD话音降噪方法

为了解决低速CVSD编码量化噪声较大、话音质量较差的问题,提出了一种基于LMS自适应滤波的CVSD话音降噪方法。首先分析了CVSD编码量化噪声的统计特性,然后在传统双通道LMS自适应滤波器的基础上,利用话音的短时相关性和量化噪声的非相关性,构造了一个采用话音延时信号作为参考输入的单通道LMS自适应滤波器。仿真实验表明,该方法可以有效滤除量化噪声,提高话音质量,具有一定的工程应用参考价值。     

量化噪声对衰落信道下MRC合并信噪比的影响

文中主要研究瑞利平衰落信道下量化噪声对最大比合并(MRC)接收分集合并输出信噪比的影响.文中推导了量化噪声与信道噪声、信道参数之间的数学关系.仿真结果阐明量化对分集度几乎没有影响,只是使系统产生了固定的信噪比恶化.无论是1%中断率对应的信噪比,还是平均信噪比,1bit量化造成的损失约为2dB,而4bit量化的损失则在0.2～0.9dB左右.     

Quantization Effects in All-Digital Phase-Locked Loops

This brief analyzes the impact of the quantization noise sources in all-digital phase-locked loops (ADPLLs), recently employed as frequency synthesizers. In general, the in-band phase noise is not only caused by the phase quantization of the time-to-digital converter, but also by the frequency quantization of the digitally controlled oscillator (DCO). The delta-sigma modulator placed at the DCO input refines the frequency quantization and adds another source of in-band PLL noise. Interestingly, the higher the modulator order, the higher this source of in-band phase noise. A method for the estimation of all the quantization noise contributors is provided, which is proven by mixed-mode simulations.     

A Quantization Noise Pushing Technique for $DeltaSigma$ Fractional- $N$ Frequency Synthesizers

This paper demonstrates our proposed quantization noise pushing technique, which moves the quantization noise to higher frequencies and utilizes the low-pass characteristic of the phased-lock loop (PLL) to further suppress the quantization noise. In addition, it can separate the operating frequency of the DeltaSigma modulator and the comparison frequency of the phase/frequency detector (PFD) so as to reduce the loop gain of the PLL and lower the in-band phase noise. This synthesizer was fabricated using the UMC 0.18-mum CMOS process. The chip area measures 0.85 mm². The supply voltage is 2 V, corresponding to a total power consumption of 26.8 mW. The experimental results show that, with this technique, the in-band phase noise can be lowered by 12 dB, while the out-of-band phase noise can be reduced by more than 15 dB, compared to a synthesizer with the same PFD comparison frequency.     

Statistics of companding noise for QAM transmission

In modern voiceband data communication, the received signal is subject to nonlinear quantization noise due to companding. Under certain conditions, this quantization noise may become dominant and cause serious degradation in performance. In order to design better signal constellations for this environment, it is of interest to characterize this noise. We study the conditional probability density function (PDF) for the quantization noise found on a PCM companding channel. We obtain an analytical expression for the conditional PDF of the nonlinear quantization noise and an algorithm to compute the conditional PDF of the filtered noise, which is the actual noise at the slicer input of the linear receiver. We compare the results to a Gaussian PDF (with second-order moments that depend on the signal point) and conclude that a Gaussian approximation, although not entirely accurate, is not unreasonable     

The quantization noise spectrum of a sinusoid in colored noise

The power density spectrum (PDS) of analog-to-digital (A/D) quantization noise is obtained for an input signal consisting of sinusoid and colored noise. While this type of quantization noise is often assumed to be uncorrelated or white, it may not be when the input signal occupies a relatively small number of quantization levels. Equations describing the PDS are derived and presented for combinations of deterministic and random A/D converter inputs     

Statistics of the Quantization Noise in 1-Bit Dithered Single-Quantizer Digital Delta–Sigma Modulators

An analysis of the quantization noise introduced by a widely-used class of single-quantizer digital delta-sigma (DeltaSigma) modulators with low-level, 1-bit dither is presented. Necessary and sufficient conditions are derived that ensure, in an asymptotic sense, various ensemble statistical properties of the quantization noise such as uniformity and independence from the input and delayed versions of itself. The conditions are also shown to be sufficient for a single realization of the quantization noise sequence to possess these properties in a time-averaged sense. Several of the most commonly-used digital DeltaSigma modulators are shown to satisfy the conditions     

On lattice quantization noise

We present several results regarding the properties of a random vector, uniformly distributed over a lattice cell. This random vector is the quantization noise of a lattice quantizer at high resolution, or the noise of a dithered lattice quantizer at all distortion levels. We find that for the optimal lattice quantizers this noise is wide-sense-stationary and white. Any desirable noise spectra may be realized by an appropriate linear transformation (“shaping”) of a lattice quantizer. As the dimension increases, the normalized second moment of the optimal lattice quantizer goes to 1/2πe, and consequently the quantization noise approaches a white Gaussian process in the divergence sense. In entropy-coded dithered quantization, which can be modeled accurately as passing the source through an additive noise channel, this limit behavior implies that for large lattice dimension both the error and the bit rate approach the error and the information rate of an additive white Gaussian noise (AWGN) channel     

Adaptive Noise Cancellation Techniques in Sigma–Delta Analog-to-Digital Converters

Adaptive noise cancellation (ANC) techniques that extract a desired signal from background noise have many applications in different engineering disciplines. In ANC, the corrupted signal is passed through a filter that tends to suppress the noise while leaving the original signal unchanged. This paper demonstrates that the adaptive noise cancellation technique can be embedded in the digital signal postprocessing of a sigma-delta analog-to-digital converter and effectively reduces the quantization noise as well as the thermal noise at the output of the converter. The combination of ANC and the noise-shaping technique enable high-resolution analog-to-digital conversion in wideband applications where noise shaping alone cannot provide enough suppression of quantization noise due to the low oversampling ratio.     

基于噪声分布特性的压缩视频超分辨率重建

该文通过对视频压缩过程进行建模,利用比特流中的量化信息和运动信息,建立了量化噪声和运动估计噪声模型,并考虑成像过程的加性噪声,总的噪声模型对不同的量化器可自适应调整。以Huber-Markov 随机场作为图像的先验模型,用梯度下降法进行MAP超分辨率重建,对其特性进行了分析。仿真实验表明,该算法重建图像的主、客观质量较高。     

Quantization Noise Suppression in Digitally Segmented Amplifiers

In this paper, we consider the problem of out-of-band quantization noise suppression in the general family of direct digital-to-RF (DDRF) conversion circuits, where the RF carrier is amplitude modulated by a quantized representation of the baseband signal. Hence, it is desired to minimize the out-of-band quantization noise in order to meet stringent requirements such as receive-band noise levels in frequency-division duplex transceivers. In this paper, we address the problem of out-of-band quantization noise by introducing a novel signal-processing solution, which we refer to as “segmented filtering (SF).” We assess the capability of the proposed SF solution by means of performance analysis and results that have been obtained via circuit-level computer simulations as well as laboratory measurements. Our proposed approach has demonstrated the ability to preserve the required signal quality and power amplifier (PA) efficiency while providing more than 35-dB attenuation of the quantization noise, thus eliminating the need for substantial post-PA passband RF filtering.