一种增量式软件开发工时度量方法研究

为了定量地管理增量式软件的迭代开发,进一步做好增量式软件开发资源分配和成本估算,需要度量开发工时.每个迭代阶段增量需求是以一定的概率实现的,把这个概率作为开发任务完成率的度量,提出增量式软件开发迭代周期的度量方法,进而获得对整个软件项目开发工时的预测.通过实例对方法进行了验证..     

Improving energy efficiency of incremental relay based cooperative communications in wireless body area networks

In this paper, we investigate the energy efficiency of an incremental relay based cooperative communication scheme in wireless body area networks (WBANs). We derive analytical expressions for the energy efficiency of direct and cooperative communication schemes taking into account the effect of packet error rate. The following communication scenarios specific to a WBAN are considered: (i) in‐body communication between an implant sensor node and the gateway, and (ii) on‐body communication between a body surface node and the gateway with line‐of‐sight (LOS) and non‐LOS channels. The results reveal a threshold behavior that separates regions where direct transmission is better from regions where incremental relay cooperation is more useful in terms of energy efficiency. It is observed that, compared with direct communication, incremental relay based cooperative communication schemes improves the energy efficiency significantly. Further, cooperation extends the source‐to‐destination hop length over, which energy efficient communication can be achieved as compared with direct communication. We also observe that, for both direct as well as cooperative transmission schemes in error prone channels, an optimal packet size exists that result in maximum energy efficiency. Copyright © 2013 John Wiley & Sons, Ltd.     

基于增量式更新带权差别集合的属性约简算法

在面向差别矩阵的约简算法思想的基础上,定义一种新的带权差别集合(WDS)模型,并提出了高效更新带权差别集合的算法,分析了该更新算法的时间和空间复杂度.随后,基于快速更新带权差别集合算法,提出一种增量式属性约简方法.当有新的数据对象被加入决策表,可有效提高属性约简的效率.理论分析和实验结果表明该算法适用于大数据集的约简.     

Incremental-based Nonlinear Detection Algorithm for MIMO System

A novel incremental nonlinear detection algorithm is presented for Multiple-Input Multiple-Output （MIMO） system. In this algorithm, the received data at multiple receiver antennas are nonlinearly mapped and then summed with weights. The weight coefficients are incrementally computed to avoid direct computation of the inverse of a matrix, which greatly reduce the computational complexity. Simulation and comparison show that the proposed algorithm can obtain better performance of Bit Error Rate （BER） than linear Minimum Mean Square Error （MMSE）.     

High Power Fiber Bundle Array Coupled LDA Module

1 Introduction High power laser diode arrays (LDA) have many advan- tages such as small volume, long working life, high slope efficiency and high optical density, so they have many applications in medical treatment, material pro- cessing, and for the pumping source of solid laser and etc. But unfortunately, the LDA can not be easy to use directly in these fields because of their poor beam quality and extremely asymmetric divergent beams (!x≈ 5°～10°and !y≈20°～35°, for example), so it …     

融合奇异值分解和线性鉴别分析的人脸识别算法

本文提出了奇异值分解（SVD）和线性鉴别分析（LDA）相结合的人脸识别算法。理论上，当两种数据或分类器具有一定的独立性或互补性时，数据融合或分类器融合才能改善识别率。SVD和LDA之间有着明显的互补之处，LDA在fisher准则下能最大限度地把不同的类别区分开来，但作为一种子空间方法，LDA敏感于位移、旋转等几何变换。而作为一种代数特征提取方法的SVD则具有位移、旋转不变性等优点。因此，将这两种方法相结合就有可能提高分类性能（好于单独的SVD方法和单独的LDA方法）。在ORL数据库上的实验表明，SVD和LDA相融合的识别方法的确提高了人脸识别率。     

Incremental POP Learning

In recently proposed partial oblique projection (POP) learning, a function space is decomposed into two complementary subspaces, so that functions belonging to one of which can be optimally estimated. This paper shows that when the decomposition is specially performed so that the above subspace becomes the largest, a special learning called SPOP learning is obtained and correspondingly an incremental learning is implemented, result of which equals exactly to that of batch learning including novel data. The effectiveness of the method is illustrated by experimental results.     

An incremental intrusion detection system using a new semi‐supervised stream classification method

In recent years, the utilization of machine learning and data mining techniques for intrusion detection has received great attention by both security research communities and intrusion detection system (IDS) developers. In intrusion detection, the most important constraints are the imbalanced class distribution, the scarcity of the labeled data, and the massive amounts of network flows. Moreover, because of the dynamic nature of the network flows, applying static learned models degrades the detection performance significantly over time. In this article, we propose a new semi‐supervised stream classification method for intrusion detection, which is capable of incremental updating using limited labeled data. The proposed method, called the incremental semi‐supervised flow network‐based IDS (ISF‐NIDS), relies on an incremental mixed‐data clustering, a new supervised cluster adjustment method, and an instance‐based learning. The ISF‐NIDS operates in real time and learns new intrusions quickly using limited storage and processing power. The experimental results on the KDD99, Moore, and Sperotto benchmark datasets indicate the superiority of the proposed method compared with the existing state‐of‐the‐art incremental IDSs.     

Fingerprint indexing based on minutiae-dependent statistical codes

Methods based on minutiae matching have been extensively used in fingerprint recognition because minutiae can be reliably extracted from poor quality and noisy fingerprints. However, structures have to be defined due to minutiae displacements and irreproducibility. Some of the structures, though very efficient, incur large computational complexities. In this article, a feature vector of statistically based values derived from the minutiae pattern in a fingerprint is proposed for indexing fingerprints using the incremental search retrieval method. The proposed indexing technique in combination with the incremental retrieval method proves to have an added advantage over certain minutiae-based structures, especially when the minutiae points are numerous in a fingerprint. The feature vector for a fingerprint requires negligible storage resources and, consequently, the computational time in the retrieval of a candidate list for a query fingerprint is very little.     

基于LDA算法的人脸识别方法的比较研究

线性判别分析（LDA）是一种较为普遍的用于特征提取的线性分类方法。但是将LDA直接用于人脸识别会遇到维数问题和“小样本”问题。人们经过研究，通过多种途径解决了这两个问题并实现了基于LDA的人脸识别。文章对几种基于LDA的人脸识别方法做了理论上的比较和实验数据的支持，这些方法包括Eigenfaces、Fisherfaces、DLDA、VDLDA及VDFLDA。实验结果表明VDFLDA是其中最好的一种方法。