基于相关性反馈的开源系统跨层需求追踪方法

大型复杂软件系统的需求分析与生成是一个由上而下逐层分解的过程,跨层需求间追踪关系的构建对于项目的管理、开发与演化都至关重要.开源系统的松耦合贡献方式需要每位参与者能便捷地理解需求的来龙去脉及需求状态,这依赖跨层需求间的追踪.问题描述日志是开源系统中需求的常见呈现方式,其无固定模板要求,内容多样(含文本、代码、调试信息等),术语使用自由,跨层需求间抽象层次大,给自动追踪带来极大的挑战.提出一种面向关键特征维度的相关性反馈方法,通过静态分析项目代码结构,抽取代码相关术语及其间的关联强度,构建代码词汇库,以缓解跨层需求的抽象层次差距及用语不统一的问题;通过度量词汇对需求描述的重要性并基于此筛选关键特征维度,以对查询语句进行针对性的优化,有效减少需求描述长度、内容形式等方面的噪音.通过在3个开源系统需求集上针对两个场景的实验,表明所提方法相比基线方法在跨层需求追踪方面的优越性,相比VSM、Standard Rocchio和Trace BERT, F2值的最大提升分别可达29.01%、7.45%和59.21%.     

Coons' surface method for formulation of finite element of plates and shells

In this paper, we apply the Coons' surface method and fitted function interpolation to fit boundary conditions of the finite element, and obtain different displacement functions of the plate and shell rectangular element, as well as the sectorial element, parallelogram element, ring element and quadrilateral element. The method is easily implemented and its geometric significance and mechanical conception are quite clear. Both computing technique and operational procedure are relatively unified. It is convenient to formulate conforming elements and high-precision elements, and also may be applied to formulate mixed and hybrid elements.     

Vector median filters

Median filters (MF) are used both to filter ‘salt and pepper’ noise from signals and images and in other signal processing applications. In this paper, an extension of the MF, the vector median filter (VMF), is introduced. As opposed to the MF, the VMF outputs for each window location a number of data elements. By adjusting the VMF parameters, the MF is obtained as a VMF special case. Just like the MF, the VMF filters impulses while simultaneously preserving step changes in a signal. The VMF's principal advantage is that it reduces the total stored data signal computation time while it produces visual outputs comparable to that of an MF. Deterministic and statistical properties of the VMF are examined. Computer-generated experimental results are also presented.     

Localized and configurable topology control in lossy wireless sensor networks

Wireless sensor networks (WSNs) introduce new challenges to topology control due to the prevalence of lossy links. We propose a new topology control formulation for lossy WSNs. In contrast to previous deterministic models, our formulation captures the stochastic nature of lossy links and quantifies the worst-case path quality in a network. We develop a novel localized scheme called configurable topology control (CTC). The key feature of CTC is its capability of flexibly configuring the topology of a lossy WSN to achieve desired path quality bounds in a localized fashion. Furthermore, CTC can incorporate different control strategies (per-node/per-link) and optimization criteria. Simulations using a realistic radio model of Mica2 motes show that CTC significantly outperforms an representative traditional topology control algorithm called LMST in terms of both communication performance and energy efficiency. Our results demonstrate the importance of incorporating lossy links of WSNs in the design of topology control algorithms.     

Gaussian-Hermite moment-based depth estimation from single still image for stereo vision

Depth information of objects plays a significant role in image-based rendering. Traditional depth estimation techniques use different visual cues including the disparity, motion, geometry, and defocus of objects. This paper presents a novel approach of focus cue-based depth estimation for still images using the Gaussian-Hermite moments (GHMs) of local neighboring pixels. The GHMs are chosen due to their superior reconstruction ability and invariance properties to intensity and geometric distortions of objects as compared to other moments. Since depths of local neighboring pixels are significantly correlated, the Laplacian matting is employed to obtain final depth map from the moment-based focus map. Experiments are conducted on images of indoor and outdoor scenes having objects with varying natures of resolution, edge, occlusion, and blur contents. Experimental results reveal that the depth estimated from GHMs can provide anaglyph images with stereo quality better than that provided by existing methods using traditional visual cues.     

Recompression effects in iris recognition

Rating a compression algorithms' performance is usually done in experimental studies, where researchers have frequently used JPEG pre-compressed data. It is not clear yet, if results of such compression experiments are reliable when conducted on pre-compressed data. To investigate this issue, we first study the impact of using pre-compressed data in iris segmentation and evaluate the relation between iris segmentation performance and general image quality metrics. In this context we propose a method to overcome potential problems in case using pre-compressed data sets cannot be avoided. As the second step, we conduct experimentation on the entire iris recognition pipeline. We find that overall, recognition accuracy results might not be entirely reliable in case of applying JPEG XR or JPEG2000 to JPEG pre-compressed data.     

Optimal topology control for balanced energy consumption in wireless networks

Hosts in wireless networks are usually powered by batteries, thus the lifetime of a network depends on the battery life of each individual host. One major solution to improve energy-efficiency is to minimize the total energy consumption. However, battery energy is a local resource, to save energy at each host and balance the energy consumption among all the hosts in the network appear to be more practical. In this paper, we prove that minimum weight incremental arborescence (MWIA) is the optimal solution for minimizing the maximum transmission power among a set of wireless nodes. We propose an algorithm that utilizes MWIA to construct a connected topology, called MWIA-based Topology Control(MWIA-TC) algorithm. We further apply MWIA-TC to the accumulated energy consumption. The theoretical analysis and experimental results show that MWIA-TC outperforms a well-known algorithm, Minimum Incremental Power (MIP), in both energy saving and network lifetime extension.     

Weakly distributive domains(Ⅱ)

In our previous work (Inform. and Comput., 2005, 202: 87–103), we have shown that for any ω-algebraic meet-cpo D, if all higher-order stable function spaces built from D are ω-algebraic, then D is finitary. This accomplishes the first of a possible, two-step process in solving the problem raised (LNCS, 1991, 530: 16–33; Domains and lambda-calculi, Cambridge Univ. Press, 1998) whether the category of stable bifinite domains of Amadio-Droste-G?bel (LNCS, 1991, 530: 16–33; Theor. Comput. Sci., 1993, 111: 89–101) is the largest cartesian closed full subcategory within the category of ω-algebraic meet-cpos with stable functions. This paper presents the results of the second step, which is to show that for any ω-algebraic meet-cpo D satisfying axioms M and I to be contained in a cartesian closed full sub-category using ω-algebraic meet-cpos with stable functions, it must not violate MI^∞. We introduce a new class of domains called weakly distributive domains and show that for these domains to be in a cartesian closed category using ω-algebraic meet-cpos, property MI^∞ must not be violated. Further, we demonstrate that principally distributive domains (those for which each principle ideal is distributive) form a proper subclass of weakly distributive domains, and Birkhoff’s M ₃ and N ₅ (Introduction to Lattices and order, Cambridge Univ. Press, 2002) are weakly distributive (but non-distributive). Then, we establish characterization results for weakly distributive domains. We also introduce the notion of meet-generators in constructing stable functions and show that if an ω-algebraic meet-cpo D contains an infinite number of meet-generators, then [D → D] fails I. However, the original problem of Amadio and Curien remains open.     

Fairness of traffic controls for inelastic flows in the Internet

In best-effort networks, fairness has been used as a criterion to guide the design of traffic controls. The notion of fairness has evolved over time, from simple equality to a form of equality modulated by the user’s need (e.g. max–min and proportional fairness). However, fairness has always been defined on a per-user basis for a deterministic workload. In this paper, we argue that we must redefine the notion of fairness when we study traffic controls for the co-existence of elastic and inelastic traffics. Our results indicate that subjecting inelastic flows to fairness congestion control on a per-flow basis does not necessarily maximize the network’s utility. Instead, inelastic flows may follow their own form of traffic control, such as admission control (without congestion control). At the aggregate level, our results indicate that it still makes sense to maintain a balance between elastic and inelastic traffic. In order to support our arguments, we develop a methodology for comparing different traffic controls for given utility functions and different workloads, both deterministic and stochastic.