基于多尺度连接模型的脑MRI模糊分类

针对传统分割方法在分割数量上的限制,以及模糊聚类方法在层次优化上的不足,提出一种有效的基于多尺度连接模型的人脑磁共振图像模糊分类算法。对脑磁共振图像进行灰度不均匀性校正后,该方法通过非线性扩散连接模型引入尺度间的模糊约束,定义2个模糊距离,并应用到模糊聚类算法中,组合尺度间和尺度内的模糊约束,给出一个多分辨模糊聚类框架。实验结果表明了该算法的准确性和有效性。     

环形激光陀螺测试技术的研究与设计

针对某环形激光陀螺传感器的总体输出误差特点,应用Allan方差法完成了对RLG频域输出误差的建模评估,包括陀螺的随机游走误差、零偏稳定性误差,以及马尔可夫噪声误差等,建立了温度误差零偏补偿模型。在陀螺测试过程中,采用了转台交差逆转的方法,尝试将零偏衰减中的常值漂移和线性漂移部分自动补消,结果表明可以改善传统的RLG转台测试法的测试精度。     

Device reliability under electrical stress and photo response of oxide TFTs

Abstract— The stability of oxide TFTs has been the main focus of this research and is probably the most crucial requirement for the successful application to flat‐panel displays. Although the high Fermi level of oxide semiconductors makes TFTs basically stable under electrical stress, the device reliability under diverse variations of electrical stress is affected by materials such as active semiconductors and gate insulators, processes for the formation of back/front channels and passivation layers, and device configurations among other things. How these factors affect the device reliability have been investigated and a review of the stability is presented. In addition, several categories of the light instability of oxide TFTs is presented and the origin is discussed.     

Cognitive-bias toward gaming-related words and disinhibition in World of Warcraft gamers

This study investigated cognitive biases toward gaming-related words and differences in cognitive performance among twelve World of Warcraft players (WWP) and thirty non-players (NP). We measured response to valenced common English and WoW jargon words using a computer-based go/no-go task. Sometimes positive valence words were the targets for the ‘go’ response, with negative-valence words as the distracters, sometimes the reverse. Target discrimination (d′) and response disinhibition (C) were calculated using a signal detection analysis. Based on questionnaire responses, there were no differences between groups in depression, anxiety, smoking or drinking, but WWP reported significantly more screen and gaming time (17.31 h/week versus 4.12 among NP). WWP had faster reaction time (RT) and better discrimination of targets from distracters (high d′) but also showed higher disinhibition (low C). WWP also showed cognitive-bias toward game-related words in the form of higher d′ for WoW jargon than common English and more disinhibition to positive-valence WoW jargon. Similar to past studies which have found alcoholics to have cognitive biases toward alcohol-related words, WWP who play frequently showed cognitive biases toward words related to the World of Warcraft game.     

<Emphasis Type="Italic">New-Age</Emphasis>: A Negative Bias Temperature Instability-Estimation Framework for Microarchitectural Components

Degradation of device parameters over the lifetime of a system is emerging as a significant threat to system reliability. Among the aging mechanisms, wearout resulting from Negative Bias Temperature Instability (NBTI) is of particular concern in deep submicron technology generations. While there has been significant effort at the device and circuit level to model and characterize the impact of NBTI, the analysis of NBTI’s impact at the architectural level is still at its infancy. To facilitate architectural level aging analysis, a tool capable of evaluating NBTI vulnerabilities early in the design cycle has been developed that evaluates timing degradation due to NBTI. The tool includes workload-based temperature and performance degradation analysis across a variety of technologies and operating conditions, revealing a complex interplay between factors influencing NBTI timing degradation.     

基于九轴MEMS定位系统误差分析与补偿算法研究

基于九轴MEMS定位系统,对由陀螺仪零偏造成的位移平面偏转误差进行分析与补偿。通过实验与推导,首次给出偏差的定量分析和误差传播的数学模型。为估计偏差,设计了一种新的状态向量中仅包含陀螺仪零偏的、简化扩展卡尔曼滤波（ SEKF）技术,克服了无加速度计辅助条件下的偏差估计问题,降低了计算量,能够很好地满足位移实时估计需求。硬件实验证明：经过偏差补偿的九轴MEMS定位系统,能够准确估计有限范围内周期性运动的位移,幅值定位精确度达到90%。该系统可用于无人机、机器人定位等领域。     

基于受约束偏置的概率矩阵分解算法

在概率矩阵分解(PMF)模型拟合之后,评分较少用户的特征趋近于先验分布的平均值,导致对其评分预测接近物品的平均评分.受约束概率矩阵分解(CPMF)未考虑到不同评分系统的整体差异以及数据集内部用户与物品存在的固有属性.针对以上问题,提出将传统矩阵分解中的用户和物品偏置项以及全局平均分结合受约束概率矩阵分解来建立新的矩阵分解算法.算法利用整体平均分衡量不同评分系统,在采用偏置来表示用户以及物品之间相互独立的属性的同时,引入约束使行为相近用户拥有相近的用户偏置,从而提高预测精度.在两个真实数据集上的实验结果表明,该算法相对于PMF和CPMF算法预测精度得到了提高.     

噪声ICA模型去偏技术研究

偏差去除算法通常假设高斯噪声条件下对普通ICA算法进行修正来消除噪声带来的影响。但是存在高斯噪声条件时,普通ICA算法对解混矩阵仍然可以辨识。故引入基于QR分解的RLS自适应噪声抵消算法和Fast-ICA算法相结合,只需对观测信号进行去噪处理,不需要对解混矩阵修正。并分别在同一噪声和相关噪声条件下做了仿真实验,与LMS-ICA算法进行了比较。仿真实验证明,该方法比后者效果显著。提出了用最小二乘算法计算分离信号的输出信噪比,作为评价算法的性能指标。     

开发可统计任意密码子用法的软件BestCodon

密码子偏好是自然现象,分析密码子的用法可以优化表达基因,即密码子优化,从而调节目的基因在特定宿主中的表达水平。现有密码子用法分析软件多数只限于单个密码子,适用于密码子对和密码子三联体的较少见。本文提出1种可统计任意密码子用法的软件结构,开发出1个密码子用法统计软件BestCodon,拥有序列检查和密码子用法统计2个模块。检查模块可代替人工校对,剔除输入序列中非规范字符,为后续分析提供正确的、符合规范的数据。统计模块根据用户指定的密码子组合类型,结果返回特定的密码子用法表。本文还探讨序列检查和密码子用法统计中计算机的运算特性,结果表明该软件检查1条长达150000 bp的序列仅耗时约0.2 s,而从中统计密码子十联体的用法仅耗时约0.1 s。