Ranked Query Processing in Uncertain Databases

Recently, many new applications, such as sensor data monitoring and mobile device tracking, raise up the issue of uncertain data management. Compared to "certain” data, the data in the uncertain database are not exact points, which, instead, often reside within a region. In this paper, we study the ranked queries over uncertain data. In fact, ranked queries have been studied extensively in traditional database literature due to their popularity in many applications, such as decision making, recommendation raising, and data mining tasks. Many proposals have been made in order to improve the efficiency in answering ranked queries. However, the existing approaches are all based on the assumption that the underlying data are exact (or certain). Due to the intrinsic differences between uncertain and certain data, these methods are designed only for ranked queries in certain databases and cannot be applied to uncertain case directly. Motivated by this, we propose novel solutions to speed up the probabilistic ranked query (PRank) with monotonic preference functions over the uncertain database. Specifically, we introduce two effective pruning methods, spatial and probabilistic pruning, to help reduce the PRank search space. A special case of PRank with linear preference functions is also studied. Then, we seamlessly integrate these pruning heuristics into the PRank query procedure. Furthermore, we propose and tackle the PRank query processing over the join of two distinct uncertain databases. Extensive experiments have demonstrated the efficiency and effectiveness of our proposed approaches in answering PRank queries, in terms of both wall clock time and the number of candidates to be refined.     

Sliding mode control of uncertain switched delay systems via hysteresis switching strategy

This paper investigates the robust sliding mode control (SMC) problem for a class of uncertain switched systems with time-varying delay. The sliding surface is constructed such that the sliding motion is completely invariant to all admissible uncertainties. For the case of the known delay-derivative upper bound, by using the multiple Lyapunov functions method, the Hysteresis switching law dependent on the state and the previous value of switching signal are designed to stabilize the sliding motion and avoid the chattering. Variable structure controllers are developed to drive the state of switched systems to reach the sliding surface in finite time and remain on it thereafter. For the case of the unknown delay-derivative upper bound, based on the single Lyapunov function method, the conditions of stabilization are obtained. Finally, a numerical example is given to illustrate the effectiveness of the proposed methods.     

北斗2代卫星导航电文纠错校验设计与仿真

提出了一种制表映射的BCH纠错解码和校验方法,通过对RNSS-ICD文档北斗导航电文BCH编码的深入理解,设计出了一种占用CPU资源少、实现简单方便的制表映射的编码和纠错解码算法,减小了算法在实现时对CPU和内存资源的占用;给出了一种再编码校验的校验方法仿真验证了该算法的准确性和可靠性,保证导航电文的可靠性;对提高接收机的性能具有很强的工程适用价值.     

统一数据管理空间在大规模并行作业计算中的应用

大规模并行作业的计算通常涉及海量的计算数据和众多的高性能计算设备.随着网格计算技术帮助人们进行计算的同时,大规模并行作业的数据规模的增长也越来越快,对计算速度的要求也越来越高,为了充分利用网格等计算平台上的资源,提高作业的计算效率,人们通常需要将待计算的数据进行分组,然后分别上传至不同的平台上进行计算,这对科学研究和数据管理造成了极大的不便.本文提出了一个针对大规模并行作业计算的统一数据管理空间,实现了异构网格和计算平台上数据的逻辑整合,从而大大提高了对计算数据的管理效率,加快了科学活动的进程.本文最后通过统一数据管理空间在大规模虚拟筛选中的应用,对该统一空间的数据传输效率和数据管理能力进行了分析.     

车载环境下基于样本熵的语音端点检测方法

在语音处理中一个关键性问题是如何准确找到语音的起止位置,目前提出许多的语音端点检测算法不能得到理想的检测结果.由于样本熵是近似熵的改进算法,提出车载环境下基于样本熵的语音端点检测方法.并采用模糊C均值聚类算法和贝叶斯信息判决算法进行样本熵特征门限估计,以及使用双门限法进行语音端点检测.在TIMIT连续语音库上的实验表明,车载噪声环境下,样本熵法和近似熵法的检测正确率均远高于谱熵法和能量谱熵法,而样本熵法相对于近似熵法具有更好的检测效果,特别是当信噪比小于等于OdB时,样本熵法的检测性能优于近似熵法近10%.因此,样本熵法在车栽智能语音领域具有很好的应用前景,能够为车载导航提供准确的语音端点检测技术.     

基于OpenGVS半潜船操作模拟器视景系统的研究

为了满足半潜船作业人员学习和培训的需要,有必要在航海模拟器大环境下研发半潜船操作模拟器仿真系统。以船舶静力学为基础,实现了半潜船装载重大件货物的过程的动态稳性计算,以Visual C++6.0和OpenGVS场景管理软件为开发平台,研制开发了半潜船操作模拟器应具备的视景系统。该仿真系统实现了视点的转移和跟踪,通过设置在半潜船的驾驶台、甲板及外部三个虚拟相机,实时动态显示半潜船各种运动的状态,同时实现了半潜船操作模拟器几个模块间的相互通信,取得了很好的效果。     

Finding the least influenced set in uncertain databases

Due to the inherent existence of uncertainty in many real-world applications, in this paper, we investigate an important query in uncertain databases, namely probabilistic least influenced set (PLIS) query, which retrieves all the uncertain objects in an uncertain database such that they are the least affected by a given query object with high probabilities. Such a PLIS query is useful in applications such as business planning. We propose and tackle both monochromatic and bichromatic versions (i.e. M-PLIS and B-PLIS, respectively) of the PLIS query. In order to efficiently answer PLIS queries, we present three pruning methods, MINMAX, Regional, and Candidate pruning, which can effectively reduce the PLIS search space. The proposed pruning methods can be seamlessly integrated into efficient query procedures. Moreover, we also study important variants of PLIS query with uncertain query object (i.e. UQ-PLIS). Furthermore, we formulate and tackle the PLIS problem on uncertain moving objects (i.e. UMOD-PLIS). Extensive experiments have demonstrated the efficiency and effectiveness of our proposed approaches under various settings.     

基于综合本体相似度计算的WEB服务发现算法

服务发现的主要过程就是对服务请求和己注册服务间IOPE进行匹配.提出一种新的服务发现框架,以及改进的综合相似度算法.综合考虑语义距离,本体树中概念的深度、密度,以及信息共享来计算服务的相似度.将服务发现过程分为过滤和排序两阶段,并对服务的各参数分别采用更适合的相似度计算方法.将用户给出的服务输入用于最后的排序阶段,以减...     

WindowsCE.NET中断延迟测试与分析

嵌入式系统对实时性能一般要求较高,中断延迟是评估嵌入式操作系统实时性能的重要指标。利用ILTiming测试工具对WinCE4.2和5.0进行中断延迟测量,分析了导致中断延迟的原因,同时针对PXA255处理器的BSP提出了测试WindowsCE系统中断延迟的方法。     

A new block-based method for copy move forgery detection under image geometric transforms

Copy move forgery detection (CMFD) is one of the most active subtopic in forgery scheme. The methods of CMFD are divided into to block-based method and keypoint-based method in general. Compared with keypoint-based method, block-based method can detect undetectable detail without morphology segmentation. But many block-based methods detect the plain copy-move forgeries only. They have been incompetent to detect the post-processing operations such as various geometrical distortions, and then fail to detect the forgery regions accurately. Therefore, this paper presents an improved block-based efficient method for CMFD. Firstly, after pre-processing, an auxiliary overlapped circular block is presented to divide the forged image into overlapped circular blocks. The local and inner image feature is extracted by the Discrete Radial Harmonic Fourier Moments (DRHFMs) with the overlapped circular block from the suspicious image. Then, the similar feature vectors of blocks are searched by 2 Nearest Neighbors (2NN) test. Euclidean distance and correlation coefficient is employed to filter these features and then remove the false matches. Morphologic operation is employed to delete the isolated pixels. A series of experiments are done to analyze the performance for CMFD. Experimental results show that the new DRHFMs can obtain outstanding performance even under image geometrical distortions.