低信噪比下运动目标的被动定位和参数估计

本语文使用两阶段的ＳＴＩ（空－时积分）定位算法实施机动目标的被动定位，第一阶段使用模拟退火算法，得到目标轨迹的初始估计；第二阶段将模拟退火算法获得的初始估计作为输入，使用修正的牛顿算法，得到目标轨迹的最后估计。计算机模拟表明，即使在低信噪比情况下，上述技术也能获得很好的目标定位效果。     

低信噪比环境下语音端点检测改进方法

针对语音端点检测在低信噪比环境下普遍存在检测性能急剧下降的问题,提出一种将调制域(时间-频率域)谱减法和自相关函数相结合的语音端点检测算法。该算法首先利用调制域谱减法较好的消噪能力来提高含噪语音的信噪比;然后根据语音和噪声的自相关函数的主峰最大值和次大值之比差异较大的特性,结合基于对数能量和自相关函数的端点检测方法对消噪后的语音进行端点检测。实验结果表明,该算法在低信噪比的环境下能取得较好的端点检测效果,并具有较好的稳健性。     

CMOS影像传感器的信噪比分析

本文使用像素密度归一化方法,对比分析了市场主流厂家生产的多系列数码相机产品中CMOS影像传感器的控噪水平。结论是：2009年尼康推出的采用瑞萨代工传感器的D3S在噪声控制方面达到了顶峰。之后的4年, CMOS控噪技术的发展似乎遇到了瓶颈,各厂商转而在提升像素密度方面大力进取。     

窄带干扰下GPS接收机捕获性能的研究

针对GPS接收机抗干扰能力强的特点．分析了窄带干扰对接收机的捕获性能的影响．在扩频理论的基础上．给出了GPSC／A码接收机信号捕获的数学模型，计算得出平均捕获时间与捕获时间方差，给出窄带干扰下的接收机信号捕获的仿真结果．结果表明．当窄带干扰功率超过一定的值时，会导致GPS接收机捕获性能的严重恶化．     

一种低信噪比环境下的语音端点检测算法

端点检测技术是语音信号处理的关键技术之一，为提高低信噪比环境下端点检测的准确率和稳健性，提出了一种非平稳噪声抑制和调制域谱减结合功率归一化倒谱距离的端点检测算法。该算法首先通过抑制非平稳噪声再采用调制域谱减消除残余噪声来提升信噪比，减少语音失真。然后再提取每帧信号的功率归一化倒谱系数，计算每帧信号与背景噪声的功率归一化倒谱距离。最后将该倒谱距离作为检测参数，采用双门限判决方法进行端点检测。实验结果表明，该端点检测算法对语音帧和噪声帧具有较好的区分性。此外，在低信噪比环境下，所提出的算法对于不同类型的噪声都具有较好的稳健性。     

OFDM-CPM系统降低PAPR新方案研究

正交频分复用-连续相位调制(OFDM-CPM)系统通过CPM相关相位状态引入记忆,从而获得比传统调制方式更好的误码性能。降低OFDM-CPM系统中的峰值平均功率比(PAPR)是关键技术之一。利用恒包络的概念提出一种新的CE-OFDM-CPM调制方案,可以把OFDM-CPM系统的PAPR降至零分贝。     

低信噪比环境下的语音识别方法研究

单通道语音信号在信噪比较大的环境下经过增强后再识别,能表现出较高的识别率。但是在低信噪比环境下,增强后语音信号的识别率急剧下降。针对此种情况,提出了一种用在识别系统前端的语音增强算法,该增强算法将采集到的带噪语音信号先使用对数最小均方误差(Logarithmic Minimum Mean Square Error,Log MMSE)提高其信噪比,然后再利用改进的维纳滤波去除噪声残留并提升语音可懂度,最后用梅尔频率倒谱系数(Mel-Frequency Cepstral Coefficients,MFCC)和隐马尔科夫模型(Hidden Markov Model,HMM)对增强后的语音信号做特征提取并识别。实验分析结果表明,该方法能有效地抑制背景噪声并减少噪声残留,显著提升低信噪比环境下语音识别的准确性。     

低信噪比下扩展目标质心的高精度计算方法

分析了扩展目标在Shack-Hartmann传感器中的成像特点。针对成像特点,提出了一种提取子光斑位置的质心计算方法。该方法应用多阈值分割和多结构元形态学去噪预处理后再进行质心计算。在信噪比低、扩展度大的情况下,能够有效地抑制噪声,减小噪声对质心计算精度的影响。在不同信噪比和扩展度下,相对于一般质心方法,该方法能够显著地提高探测精度。当信噪比为0.5,扩展度达13时,该方法的平均绝对误差不到1个像素。     

高斯势分段双稳随机共振在不同噪声下的轴承故障诊断EI北大核心CSCD

为解决经典双稳随机共振(classical bistable stochastic resonance, CBSR)在强噪声下输出信噪比(signal-to-noise ratio, SNR)较低问题,将非饱和分段双稳随机共振(unsaturated piecewise bistable stochastic resonance, UPBSR)与高斯势(gaussian potential, GP)随机共振结合得到一种高斯势分段双稳随机共振(Gaussian potential piecewise bistable stochastic resonance, GPPBSR)。首先,将GPPBSR与CBSR、UPBSR的势函数分析对比;其次,以SNR和平均信噪比增益(mean signal-to-noise ratio increase, MSNRI)为衡量指标,分别在高斯白噪声和色噪声背景下通过遗传算法参数寻优得到SNR和MSNRI随系统参数的变化规律,其结果表明在两种噪声背景下,GPPBSR系统的SNR和MSNRI均优于UPBSR,抗噪性能更好;最后,为验证在不同场景下GPPBSR系统轴承故障诊断的实用性,将UPBSR和GPPBSR系统应用于6205-2RS JEM SKF和HRB 6205-2Z两种型号的轴承故障诊断中。仿真结果表明,GPPBSR系统在两种噪声背景下的故障信号诊断均是可行的,且比CBSR和UPBSR的性能更优越。     

一种低信噪比下的宽带扩频信号捕获跟踪方法

为了解决低信噪比下宽带扩频信号的捕获算法复杂、实现成本高和受多普勒频率影响性能下降的问题,该文提出了一种易于实现的并行捕获及跟踪方法。采用分段并行频域捕获方法进行初始捕获,并复用相同的算法模块对信号进行定时跟踪,同时估算出多普勒频偏补偿捕获的信号,达到在极低信噪比和存在一定的多普勒频偏条件下对连续扩频信号稳定捕获跟踪的效果,且该方法实现简单、硬件资源占用较少,已应用于工程建设中。     

低信噪比环境下语音识别的鲁棒性方法研究

针对噪声环境下语音识别率急剧下降的问题，提出了一种基于语音时频域稀疏性原理的改进最小方差无畸变响应波束形成与改进维纳滤波结合的算法。该算法首先利用麦克风阵列语音信号的空间信息，通过基于时频掩蔽的改进最小方差无畸变响应波束形成器，增强目标声源方向的语音信号，抑制其他方向噪声的干扰，然后再使用改进的维纳滤波器去除残留噪声并提高语音可懂度，对增强后的语音信号提取梅尔频率倒谱系数作为特征参数，使用隐马尔可夫模型搭建语音识别系统。实验结果表明，该方法能够有效提高低信噪比环境下的语音识别率，具有较好的鲁棒性。     

环保型低温制冷系统的性能实验

对环保型R290/CO2复叠式低温制冷系统的性能进行实验,得出R290循环的COP要比CO2循环的COP高,CO2压缩机的吸气温度对CO2压缩机排气温度的影响较明显.CO2低温循环中,随着温度的降低,制冷工质的粘性对管路的流动阻力损失影响不大.制冷系统的压力和流量的稳定性非常好,温度的稳定性能够满足实验精度要求.     

低信噪比下重构协方差矩阵的高分辨MUSIC算法

在满足对称分布的海洋噪声场中,为提高低信噪比条件下目标方位估计性能,提出一种重构信号协方差矩阵的MUSIC算法。利用数据协方差矩阵虚部与对称噪声无关的性质,根据协方差矩阵虚部和虚部MUSIC算法的预估角重构出信号协方差矩阵,在此基础上实现MUSIC算法。仿真结果表明,所提算法相比常规MUSIC算法能有效降低对称噪声的影响,提高方位估计性能,并避免双边谱的出现,有更高的分辨率和更低的分辨门限。还研究了协方差矩阵的Toeplitz修正处理对于MUSIC类算法的改善作用。仿真表明,Toeplitz修正处理能显著提高MUSIC类算法的分辨性能。     

Performance of ejector cooling systems using low ecological impact refrigerants

The theoretical behaviour of an ejector cooling system, using as working fluids propane, butane, isobutane, R152a and R134a, is obtained. The ejector works as a thermo-compressor that is simulated with a validated one-dimensional mathematical model, whose errors are lower than 6%. For a system unitary cooling capacity, a parametric study is carried out varying the generation, condensation and evaporation temperatures. From the obtained data, a complete analysis of the system performance can be achieved when the ejector and system operation parameters are considered. The best performance corresponds to the system using propane, because has the highest system coefficient of performance and its ejector has the maximum entrainment ratio value, the least area ratio value and the highest efficiency value. The considered generation temperature ranging from 70 °C to 95 °C is appropriate for low-grade energy sources assisting thermal cooling systems. After this system performance, come those in which R152a and R134a are employed, with isobutane and butane at the end. The obtained results represent potential design points of an efficient ejector cooling system operation, because to each combination of the above mentioned temperatures corresponds one and only one ejector geometry.     

Comparative study of push and pull systems considering quality performance in a cell-based job shop environment

Simulation experiments were used to investigate the influences of quality performance on the performance difference between push and pull systems in a cell-based job shop environment. Quality performance was represented by two factors, i.e. mean magnitude of step shifts (MMSS) and mean frequency of step shifts (MFSS). Shop performance was based on the average and standard deviation of flow time. Results show that push systems are superior to pull systems, regardless of quality performance and set-up time reduction (STR) effected by cellular manufacturing. Therefore, to justify the adoption of pull systems, their potentials beyond material flow control (e.g. in promoting continuous improvement) need to be realized. In addition, the larger the STR, the more effective the quality performance improvement. With a large STR (of 80%), reduction of MMSS or (and) MFSS can improve the average (standard deviation) of flow time. In particular, reduction of MFSS tends to be more effective for a pull system with a larger STR. Accordingly, for a pull system in a cell-based job shop environment, substantial set-up time reduction is critical not only for efficient process operations, but also for effective quality performance improvement.     

Performance measurement in a theory of constraints environment

There is an increasing recognition of the critical role played by an organization's performance measurement system in achieving competitive success. The success of an organization may well depend upon the compatibility between the performance measurement system in operation at subordinate organizational levels and the organization's global goals. An increasing amount of research in operations management addresses the relationship between the operation's performance measurement system and its apparent incompatibility with global organizational goals. The performance measurement system associated with the theory of constraints (TOC) has been advocated by some researchers as a mechanism for linking operational objectives to the global goals of an organization. However, empirical research is lacking to determine and test the relevance of the proposed performance measurement system presented in TOC. This study examines the application of a TOC-based performance measurement system in an operating environment and reports on the results. A set of propositions are also provided to further future research.     

Simulation of particle agglomeration and fragmentation in a low gravity environment

We are developing simulation tools to design aerosol experiments in a space environment. The simulations will be used to assess trade-offs among the level of gravity, active experimental volume, particle density and primary particle size. In our previous work [1] we simulated the formation of particle clusters in a low gravity environment using a modified version of the CONTAIN code [2]. The model did not include the effect of cluster fragmentation on the particle size distribution and deposition on the cell walls. Before implementing complex modifications to the CONTAIN code we have investigated the effects of fragmentation using the Monte Carlo (MC) method to simulate aerosol dynamics. The constant-N Monte Carlo method developed by the Matsoukas group [3–5] was used. In this method the number of particles in the system is kept constant and the mean cluster volume is increased or decreased depending on the event (fragmentation, agglomeration,…). We are particularly interested in measuring the average volume of clusters and probability, Pk, of finding a cluster containing k primary particles. Both quantities are measurable in a low microgravity experiment. Results are presented for different levels of gravity, particle density and fragmentation kernel. The results demonstrate the potential benefits of a full simulation of the planned experiments.     

一种适用于浅海小信噪比条件的声源目标定位技术

文章将匹配场处理技术引入一种新的目标运动分析方法——长时间积分法中，用根据实际海洋传播条件计算所得的声场作为拷贝场向量，替代传统方法中所采用的平面波模型。由于充分利用了声场的空间结构，使得该方法不仅能够完成传统目标运动分析中定出目标方位的功能，而且还可以定出其距离和深度。同时，因积分时间的增加还大大提高了对目标的检测性能。该方法可用于解决小信噪比条件下，声源目标的检测和定位问题。尤其适用于浅海、在声信号的传播因受波导和多途现象影响而严重偏离平面波的情况。仿真结果表明：采用平面波模型时，在信噪比低于-28dB时，声源目标参数就已相当模糊；而当采用匹配场处理技术，在信噪比降为-33dB时，仍可精确地得到声源目标的参数。