自适应相干噪声衰减技术

去除噪声是必需的地震数据处理过程。由于地表条件和地震波激发、传播以及接收条件等诸多方面的原因，时常引起地震记录上出现各种相干干扰波。本文主要针对地震数据处理中地震剖面上存在的各种相干噪声，提出了时问一空间域自适应相干噪声衰减技术，以进行信噪分离。这种方法把噪声在整个处理平面上的视速度近似看作常量(正值或负值)，可以有效地消除地震剖面上的线性噪声、面波以及多次波等，并且对信号的保真度高。本文方法是复杂地区低信噪比地震数据处理的一种理想的噪声去除方法，具有运算速度快、压制噪声效果好等优点。     

机载红外搜索跟踪系统被动定位滤波算法研究

首先用扩展卡尔曼滤波算法构建了机载红外搜索跟踪系统被动定位滤波模型．然后针对该滤波算法要求先验的噪声统计及存在系统观测模型线性化误差影响滤波精度的特点．利用虚拟噪声技术，提出了适合于红外搜索跟踪系统被动定位的自适应扩展卡尔曼滤波算法。该算法实时地估计了虚拟噪声的统计特性，减小了线性化误差，提高了非线性滤波的精度。仿真结果表明，在完全相同的初始条件下，自适应扩展卡尔曼滤波对目标距离和速度的估计结果明显优于扩展卡尔曼滤波，此算法具有很高的工程应用价值。     

探测激光超声信号的Fabry—Perot干涉仪及测试系统的研制

本文自行研制了一台Fabry—Perot干涉仪及其用于探测超声信号的测试系统。利用该系统对非光学表面的铝板和铜板实现了nm级窄脉冲激光超声信号的非接触式检测,其信噪比足以满足进一步信号处理要求。     

笼型感应电动机电磁噪声及其控制

本文首先讨论了笼型感应电动机的电磁噪声和负载噪声的关系及特点，然后论述了控制电磁噪声的若干措施．并首次采用了几种目前不被人们采用和推荐采用的槽配合．设计和制造了低噪声笼型三相感应电动机．给出了电磁噪声的计算值和空载噪声试验结果．     

非线性不确定大系统的分散滑动模控制

本文研究一类非线性不确定大系统的分散滑动模控制问题．首先给出了使滑动模具有良好动态品质的分散切换流形的设计方法；然后给出了一种基于不确定因素已知变化范围的分散滑动模控制器设计方法，该方法只需对各孤立子系统进行降阶滑动模的极点配置及求解降阶Ｌｙａｐｕｎｏｖ矩阵方程即可；最后以一具体例子说明其设计过程。     

激光与DNA作用系统的随机共振研究

建立了激光与ＤＮＡ分子系统相互作用的Ｆｏｋｋｅｒ－Ｐｌａｎｃｋ方程，通过对该方程的数值研究，发现在绝热近似下，系统发生随机共振，噪声强度，激光振幅和频率协同作用共同制约着生物系统的演化过程，噪声强度在生物系统遗传变异过程中可能起重要作用。     

Electrochemical noise analysis for estimation of corrosion rate of carbon steel in bicarbonate solution

We have investigated the electrochemical noise behavior of carbon steel in fully deaerated aqueous bicarbonate solutions, and discussed the optimum conditions of the noise analysis for estimating corrosion rate of the steel. Noise of the potential difference and of the short-circuit current between two identical steel coupons were successfully measured. The time-series noise patterns were transformed into frequency domain by fast Fourier transformation, and then their power spectrum densities (PSDs) at a frequency were determined to be compared with the corrosion rate. The PSDs of the potential and of the current varied with changing environmental factors of bicarbonate concentration, pH, and immersion time. The factors also controlled the corrosion rate of the steel. The PSDs were associated with the corrosion rate, and then it was found that the PSDs of the potential and of the current showed linear correlation with the corrosion rate in log-log scale. There was also linear relationship between the corrosion rate and a spectral noise resistance obtained from the PSDs of the potential and the current. The linearities of the three correlations were better at a lower analyzed frequency. Furthermore, the PSDs of the current and the noise resistance indicated more linear correlation with the corrosion rate than that of the potential. As the simplicity of the measurement system is additionally considered, it is concluded that the PSD of the current noise at an analyzed frequency of 3 mHz is the optimum conditions for estimating the corrosion rate from 10⁻² to 10⁰ A m⁻² in this study.     

Ranking with uncertain labels and its applications

The techniques for image analysis and classification generally consider the image sample labels fixed and without uncertainties. The rank regression problem studied in this paper is based on the training samples with uncertain labels, which often is the case for the manual estimated image labels. A core ranking model is designed first as the bilinear fusing of multiple candidate kernels. Then, the parameters for feature selection and kernel selection are learned simultaneously by maximum a posteriori for given samples and uncertain labels. The provable convergency Expectation Maximization (EM) method is used for inferring these parameters in an iterative manner. The effectiveness of the proposed algorithm is finally validated by the extensive experiments on age ranking task and human tracking task. The popular FG-NET and the large scale Yamaha aging database are used for the age estimation experiments, and our algorithm outperforms those state-of-the-art algorithms ever reported by other interrelated literatures significantly. The experiment result of human tracking task also validates its advantage over conventional linear regression algorithm. A short version of this paper appeared in ICME07.     

基于信息理论的采样成像系统优化设计方法

本文把信息论应用于采样成像系统中，用端到端的互信息量评价图像捕捉系统，研究光电一体化系统优化设计方法。模拟系统设计参数，计算机仿真分析了采样成像系统中普遍存在的欠采样噪声、光学成像系统的模糊效应以及二者之间的涨落、平衡，做出了基于互信息量的系统评价仿真曲线，提出了光学成像系统与图像捕捉系统的匹配条件和优化设计方法。     

轿车车内减振降噪控制方法的研究

本文分析了轿车车内噪声产生的机理，并对主动控制方法在汽车减振降噪领域的应用作了探讨和展望。