A practical classification-rule for software-quality models

A practical classification rule for a SQ (software quality) model considers the needs of the project to use a model to guide targeting software RE (reliability enhancement) efforts, such as extra reviews early in development. Such a rule is often more useful than alternative rules. This paper discusses several classification rules for SQ models, and recommends a generalized classification rule, where the effectiveness and efficiency of the model for guiding software RE efforts can be explicitly considered. This is the first application of this rule to SQ modeling that we know of. Two case studies illustrate application of the generalized classification rule. A telecommunication-system case-study models membership in the class of fault-prone modules as a function of the number of interfaces to other modules. A military-system case-study models membership in the class of fault-prone modules as a function of a set of process metrics that depict the development history of a module. These case studies are examples where balanced misclassification rates resulted in more useful and practical SQ models than other classification rules     

Using regression trees to classify fault-prone software modules

Software faults are defects in software modules that might cause failures. Software developers tend to focus on faults, because they are closely related to the amount of rework necessary to prevent future operational software failures. The goal of this paper is to predict which modules are fault-prone and to do it early enough in the life cycle to be useful to developers. A regression tree is an algorithm represented by an abstract tree, where the response variable is a real quantity. Software modules are classified as fault-prone or not, by comparing the predicted value to a threshold. A classification rule is proposed that allows one to choose a preferred balance between the two types of misclassification rates. A case study of a very large telecommunications systems considered software modules to be fault-prone, if any faults were discovered by customers. Our research shows that classifying fault-prone modules with regression trees and the using the classification rule in this paper, resulted in predictions with satisfactory accuracy and robustness.     

A comparative study of pattern recognition techniques for qualityevaluation of telecommunications software

The extreme risks of software faults in the telecommunications environment justify the costs of data collection and modeling of software quality. Software quality models based on data drawn from past projects can identify key risk or problem areas in current similar development efforts. Once these problem areas are identified, the project management team can take actions to reduce the risks. Studies of several telecommunications systems have found that only 4-6% of the system modules were complex [LeGall et al. 1990]. Since complex modules are likely to contain a large proportion of a system's faults, the approach of focusing resources on high-risk modules seems especially relevant to telecommunications software development efforts. A number of researchers have recognized this, and have applied modeling techniques to isolate fault-prone or high-risk program modules. A classification model based upon discriminant analytic techniques has shown promise in performing this task. The authors introduce a neural network classification model for identifying high-risk program modules, and compare the quality of this model with that of a discriminant classification model fitted with the same data. They find that the neural network techniques provide a better management tool in software engineering environments. These techniques are simpler, produce more accurate models, and are easier to use     

Classification-tree models of software-quality over multiplereleases

This paper presents an empirical study that evaluates software-quality models over several releases, to address the question, “How long will a model yield useful predictions?” The classification and regression trees (CART) algorithm is introduced, CART can achieve a preferred balance between the two types of misclassification rates. This is desirable because misclassification of fault-prone modules often has much more severe consequences than misclassification of those that are not fault-prone. The case-study developed 2 classification-tree models based on 4 consecutive releases of a very large legacy telecommunication system. Forty-two software product, process and execution metrics were candidate predictors. Model 1 used measurements of the first release as the training data set; this model had 11 important predictors. Model 2 used measurements of the second release as the training data set; this model had 15 important predictors. Measurements of subsequent releases were evaluation data sets. Analysis of the models' predictors yielded insights into various software development practices. Both models had accuracy that would be useful to developers. One might suppose that software-quality models lose their value very quickly over successive releases due to evolution of the product and the underlying development processes. The authors found the models remained useful over all the releases studied     

Count Models for Software Quality Estimation

Identifying which software modules, during the software development process, are likely to be faulty is an effective technique for improving software quality. Such an approach allows a more focused software quality & reliability enhancement endeavor. The development team may also like to know the number of faults that are likely to exist in a given program module, i.e., a quantitative quality prediction. However, classification techniques such as the logistic regression model (lrm) cannot be used to predict the number of faults. In contrast, count models such as the Poisson regression model (prm), and the zero-inflated Poisson (zip) regression model can be used to obtain both a qualitative classification, and a quantitative prediction for software quality. In the case of the classification models, a classification rule based on our previously developed generalized classification rule is used. In the context of count models, this study is the first to propose a generalized classification rule. Case studies of two industrial software systems are examined, and for each we developed two count models, (prm, and zip), and a classification model (lrm). Evaluating the predictive capabilities of the models, we concluded that the prm, and the zip models have similar classification accuracies as the lrm. The count models are also used to predict the number of faults for the two case studies. The zip model yielded better fault prediction accuracy than the prm. As compared to other quantitative prediction models for software quality, such as multiple linear regression (mlr), the prm, and zip models have a unique property of yielding the probability that a given number of faults will occur in any module     

Optimal design of large software-systems using N-versionprogramming

Fault tolerant software uses redundancy to improve reliability; but such redundancy requires additional resources and tends to be costly, therefore the redundancy level needs to be optimized. Our optimization models determine the optimal level of redundancy within a software system under the assumption that functionally equivalent software components fail independently. A framework illustrates the tradeoff between the cost of using N-version programming and the improved reliability for a software system. The 2 models deal with: a single task, and multitask software. These software systems consist of several modules where each module performs a subtask and, by sequential execution of modules, a major task is performed. Major assumptions are: 1) several versions of each module, each with an estimated cost and reliability, are available, 2) these module versions fail independently. Optimization models are used to select the optimal set of versions for each module such that the system reliability is maximized and total cost remains within budget     

软件方案选择灰局势决策

在分析软件方案选择影响因素的基础上，将软件方案选择看成是一次灰局势决策。提出了运用灰局势决策对方案进行选择的方法，并在计算统一测度时引入目标权重向量，得到更为合理的结果，从而实现方案的量化选择。实例表明，该方法具有很好的可信度，为软件方案选择提供了一种有效的方法。     

多目标猫群优化算法支持下的云计算任务调度

针对云环境下任务调度易出现多目标冲突的问题,提出一种改进的基于猫群的多目标优化算法。该算法模拟猫的行为模式,采用基于线性混合比率的猫行为选择方式来提高全局搜索和局部寻优能力;并在迭代过程中结合任务完成时间和任务费用支出,引入一个可调节的多目标集成效用函数,实现了资源与任务的智能调度。实验结果表明,所提算法不仅求解质量高,且在求解速度和调度消耗方面均优于多目标遗传算法和多目标粒子群算法。     

一种基于边缘分布估计的多目标优化算法

该文提出了一种基于边缘分布估计的多目标优化算法,通过在每一进化代中估计较优个体的边缘概率分布来引导算法对Pareto最优解的搜索。通过与基于拥挤机制的多样性保持技术、基于非支配排序的联赛选择、精英保留等技术的有机结合,使得算法在具有良好收敛性能的同时,具有很好的维持群体多样性的能力。通过一组典型测试函数实验对该算法的性能进行了分析,并与NSGA-II、SPEA、PAES等知名多目标优化算法进行了比较,结果表明该文算法收敛速度较快,且得到的非支配解集分布均匀,适合于复杂多目标优化问题的求解。     

Managing Test-Procedures to Achieve Reliable Software

Quantitative decision-making procedures are proposed to aid software project managers to manage effectively the testing stage during software project development. The module and integration testing phases are thoroughly investigated. Decision procedures which maximize the reliability and/or minimize some cost-benefit objective subject to a time and/or budget constraint are suggested. These procedures optimally allocate test time to the modules for module testing and select the optimal data mixture for integration testing. Testing of computer software is a major component of the software development effort. An efficient allocation of computer time among various modules during testing can appreciably improve reliability and shorten the testing stage. Using decision models presented in this paper, a project manager can effectively allocate test time during module testing and select the best data mixture for integration testing. The models are based upon software failure data that are collected during testing. These decision models can be valuable not only for the project manager but for the group responsible for generating the appropriate test data.     

融合面积估算和多目标优化的硬件任务划分算法

针对可重构计算机系统配置次数(划分块数)的最小化问题,提出了一种融合面积估算和多目标优化的硬件任务划分算法。该算法每次划分均进行硬件资源面积的估算,并且通过充分考虑可重构资源的使用、一个数据流图所有划分块执行延迟总和、划分模块间边数等因素构造了新的探测函数prior_assigned(),该函数能够计算每个就绪节点的优先权值,新算法通过该值能动态调整就绪列表任务节点的调度次序。实验结果表明,与现有的层划分、簇划分、增强静态列表、多目标时域划分、簇层次敏感等5种划分算法相比,该算法能获得最少的模块数,并且随着可重构处理单元面积的增大,除层划分算法之外,其执行延迟的均值也是最小的。     

Optimal allocation and control problems for software-testingresources

Two kinds of software-testing management problems are considered: testing-resource allocation to best use specified testing resources during module testing, and a testing-resource control problem concerning how to spend the allocated amount of testing-resource expenditures during it. A software reliability growth model based on a nonhomogeneous Poisson process is introduced. The model describes the time-dependent behavior of software errors detected and testing-resource expenditures spent during the testing. The optimal allocation and control of testing resources among software modules can improve reliability and shorten the testing stage. Based on the model, numerical examples of these two software testing management problems are presented     

基于分解的多目标进化算法的异构无线网络业务接入控制

异构无线网络接入控制问题包含多个优化目标,现有算法考虑不全面且多是将其转换为单目标求解,限制了各目标的相对关系,无法适应不同的实际需求。该文提出一种直接采用多目标进化算法的接入控制算法。首先将优化目标扩展为3个,分别是最小化阻塞率、最小化占用总资源和负载均衡;其次引入基于分解的多目标进化算法(MOEA/D)并设计进化策略,进行初步寻优;最后通过非支配排序得到Pareto最优解集,即最佳接入方案。仿真结果表明,所提算法可以提高各优化目标的求解精度,从而提高业务接入成功率和网络资源利用率,并且为决策者提供多种接入方案,可根据实际需求进行最优选择。     

基于遗传算法-响应曲面方法的激光透射焊接聚碳酸酯工艺的多目标优化

采用旋转中心复合设计进行激光透射焊接热塑性聚合物的试验规划,运用响应曲面法建立焊接工艺参数和焊接强度、焊缝宽度的关系模型,采用满意度函数和遗传算法相结合,对焊接强度、焊缝宽度和焊接成本进行了多目标优化及优化结果的验证。结果表明,优化的预测结果和实验结果取得了较好的一致性,可以为提高焊接质量和降低生产成本提供有效的指导。     

A software defined radio modeling framework for enabling cognitive applications

The rapid advances in software defined radio (SDR) technologies over the past decade encouraged their transition from the lab to field deployment, and shifted the usage from disparate communication devices to the creation of large resilient wireless networks capable of adapting to highly variable environment conditions in a cognitive manner. Simulation models capable of accurately representing the behavior and properties of SDR devices operating in a networked environment thus become a prerequisite for both the evaluation of such networks as well as for providing a development platform for creating new cognitive capabilities. This paper describes our work in creating a network simulation model framework for software defined radios that takes into account some of the unique behaviors and requirements of SDRs not previously seen in purely hardware devices. Factors such as large and variable communication delays between software modules, contention to shared resources, such as CPU or memory, as well as continuous tuning and environment awareness functionality essential to SDR demand a different modeling approach as well as novel techniques that enable accurate scale testing.     

一种非均匀分布问题分布性维护方法

 几乎所有多目标进化算法(multi-objective optimization evolutionary algorithm,MOEA)都是针对Pareto最优面为均匀分布问题而言.然而现实中很多问题Pareto最优面是非均匀分布的,决策者希望得到一个与Pareto最优面分布类似的解集.现存算法并不能有效解决该问题.对此,提出一种针对于非均匀分布多目标优化问题的维护方法(non-uniformly diversity maintenance method,NUDMM).该方法定义一个反映个体分布"规则"程度的指标——杂乱度,并设计一种降低种群杂乱度的方法,在未知Pareto最优面分布规律情况下有效剔除造成种群混乱的个体.通过与NSGA-II和SPEA2在不同维数下8个非均匀函数上对比实验,表明NUDMM在有效保持问题真实分布的同时,具有良好的收敛性.     

Radio resource management with QoS guarantees for LTE-A systems: a review focused on employing the multi-objective optimization techniques

The increasing number of subscribers’ demand has led to the evolution of future wireless networks that support multimedia applications and require ensuring the quality of services it provides. As the radio resource is becoming scarce, it is turning out to be a vital issue that how should the demands for higher data rates with limited resources is met for the evolving long term evolution-advanced (LTE-A) systems. Moreover, the efficiency and performance of resource management can be further improved by autonomously assigning and managing resources among various users and applications. We have surveyed various radio resource management (RRM) techniques being used for resource sharing in LTE-A networks that focus on the potential of multi-objective optimization algorithms for achieving desired QoS in LTE-A system. In this paper, we present a comprehensive review of RRM techniques, scheduling, and QoS along with a focus on implementing the multi-objective optimization techniques for efficient resource allocation.     

Multiproject staff scheduling with variable resource constraints

An analytical approach and a PC software based solution are presented to a project management problem, in which multiple resources must be allocated to perform both routine tasks and a set of projects with varying levels of priority. The performance of a priority-based decision rule was tested on this project management problem. The model was able to provide division-level management with an effective tool for staff workload planning and program management. A sequential, iterative approach is developed and used to test and evaluate alternative PC commercial software on a set of projects and evolving criteria. A PC commercial software package, with associated heuristics, is demonstrated to provide useful results to multiproject staff scheduling problems in an actual organization     

基于合作模型的协同免疫多目标优化算法

本文针对多目标优化问题Pareto最优解集合(PS)的分布特点,构造了一种基于新的子任务划分方法的合作型协同进化模型,并将该模型引入人工免疫系统中,提出了一种基于合作模型的协同免疫多目标优化算法(A Cooperative Immune Coevolutionary Algorithm for Multiobjective Optimization,CICAMO).CICAMO算法运用Tchebycheff分解方法进行子种群划分,然后对各个子种群建立线性概率统计模型分段逼近整个PS,在抗体繁殖上结合了克隆选择和模型采样两种方式.实验结果表明,CICAMO算法在求解质量和收敛速度上均表现良好,尤其对于决策变量非线性相关的多目标优化问题,性能尤为突出.     

Efficient classification and codebook design for CVQ

An efficient block classification and a new codebook design algorithms for classified vector quantisation (CVQ) are presented. The classification method is particularly simple and suited for hardware or software implementation. For each nonshade block, a local threshold is computed and used to binarise the block. The binary block is then compared with a predefined set of binary edge templates to decide its nature. On the other hand, the codebook design algorithm generates the codevectors in a short time, while providing considerable computational savings. It only uses the AC energy of the training set vectors to produce the codebook and the AC energy of an input block for its encoding. Subsequently, the actual codebook does not have to be present at the transmitter end. Good quality images, as measured by subjective and objective measures, are produced at bit rates in the range 0.64-0.87 bits per pixel (bpp)