并发反应式系统的组合模型检验与组合精化检验

模型检验和精化检验是两种重要的形式验证方法,其应用的主要困难在于如何缓解状态爆炸问题.基于分而治之的思想进行组合模型检验和组合精化检验是应对这个问题的重要方法,它们利用系统的组合结构对问题进行分解,通过对各子系统性质的检验和综合推理导出整个系统的性质.在一个统一的框架下对组合模型检验和组合精化检验作了系统的分析和归纳,从模块检验的角度阐述了上述两种组合验证方法的原理及其相应的组合验证策略.同时总结了各类问题的复杂性,并对上述两种方法作了比较分析,揭示了它们之间的内在联系.最后展望了组合模型检验与组合精化检验的发展方向.     

Investigating a file transfer protocol using CSP and B

In this paper a file transmission protocol specification is developed using the combination of two formal methods: CSP and B. The aim is to demonstrate that it is possible to integrate two well established formal methods whilst maintaining their individual advantages. We discuss how to compositionally verify the specification and ensure that it preserves some abstract properties. We also discuss how the structure of the specification follows a particular style which may be generally applicable when modelling other protocols using this combination .     

一种接口自动机的组合精化检验方法

接口自动机是一个用来描述软构件接口的时态行为的形式模型，传统的简单组合精化检验规则由于没有考虑到环境时子任务的影响而使其实际应用受到较大限制。本文提出了一种对该规则的改进方法，以弥补上述缺陷。     

UML模型中并发对象的组合验证

本文在用层次自动机结构化表示UML Statecharts的基础上,定义了UML协同图中并发对象的同步合成,然后根据结构间的模拟关系,研究了对并发对象系统进行组合验证的方法和规则,使有可能在对UML协同图进行模型检验的过程中不必建立系统的全局状态图,以缓解状态爆炸问题。     

基于进程迹的CSP模型验证框架

CSP(Communicating Sequential Processes)是构建并发系统和网络安全协议的经典方法。当前主流的CSP模型验证方法需将进程转化为迁移系统,转化过程比较复杂;性质采用迹进行规范,不利于活性的描述。提出了一种基于进程迹的CSP模型验证框架,其性质采用通用的规范方法LTL进行描述。利用ASP(Answer Set Programming)技术实现了一个CSP验证系统。实验表明,与类似系统相比,该系统的描述能力更强,验证结果的准确性更高,在性质不满足时还可提供反例。     

Design and Verification of Distributed Recovery Blocks with CSP

A case study on the application of Communicating Sequential Processes (CSP) to the design and verification of fault-tolerant real-time systems is presented. The distributed recovery block (DRB) scheme is a design technique for the uniform treatment of hardware and software faults in real-time systems. Through a simple fault-tolerant real-time system design using the DRB scheme, the case study illustrates a paradigm for specifying fault-tolerant software and demonstrates how the different behavioural aspects of a fault-tolerant real-time system design can be separately and systematically specified, formulated, and verified using an integrated set of formal techniques based on CSP.     

基于模拟关系的精化检测方法

精化检测是一种重要的形式化验证方法,将系统实现和性质规约用相同形式化语言进行建模,如能证明两者间存在某种精化关系且该关系能够维持性质,可得出系统实现满足性质规约.为验证不同类型的系统性质, traces、stable failures和failures-divergence精化检测方法已被提出.精化检测算法依赖于子集构造,因而其面临状态空间爆炸问题.近年来,已有学者针对NFA语言包含问题提出了基于模拟关系的状态空间消减方法,大大提高了算法性能,且该方法能直接用于traces精化检测.在此基础上,本文提出了基于模拟关系的stable failures和failures-divergence精化检测方法.此外,本文还将精化检测扩展到了时间系统的验证中,提出了基于模拟关系的时间自动机traces精化检测方法.实验结果表明,基于模拟关系的算法效率有很大提高.     

Structural Refinement of Systems Specified in Object-Z and CSP

This paper is concerned with methods for refinement of specifications written using a combination of Object-Z and CSP. Such a combination has proved to be a suitable vehicle for specifying complex systems which involve state and behaviour, and several proposals exist for integrating these two languages. The basis of the integration in this paper is a semantics of Object-Z classes identical to CSP processes. This allows classes specified in Object-Z to be combined using CSP operators. It has been shown that this semantic model allows state-based refinement relations to be used on the Object-Z components in an integrated Object-Z/CSP specification. However, the current refinement methodology does not allow the structure of a specification to be changed in a refinement, whereas a full methodology would, for example, allow concurrency to be introduced during the development life-cycle. In this paper, we tackle these concerns and discuss refinements of specifications written using Object-Z and CSP where we change the structure of the specification when performing the refinement. In particular, we develop a set of structural simulation rules which allow single components to be refined to more complex specifications involving CSP operators. The soundness of these rules is verified against the common semantic model and they are illustrated via a number of examples.     

Checking Finite Traces Using Alternating Automata

Alternating automata have been commonly used as a basis for static verification of reactive systems. In this paper we show how alternating automata can be used in runtime verification. We present three algorithms to check at runtime whether a reactive program satisfies a temporal specification, expressed by a linear-time temporal logic formula. The three methods start from the same alternating automaton but traverse the automaton in different ways: depth-first, breadth-first, and backwards, respectively. We then show how an extension of these algorithms, that collects statistical data while verifying the execution trace, can be used for a more detailed analysis of the runtime behavior. All three methods have been implemented and experimental results are presented.     

Compositional Verification of Knowledge-Based Task Models and Problem-Solving Methods

In this paper a compositional verification method for task models and problem-solving methods for knowledge-based systems is introduced. Required properties of a system are formally verified by deriving them from assumptions that themselves are properties of sub-components, which in their turn may be derived from assumptions on sub-sub-components, and so on. The method is based on properties that are formalized in terms of temporal semantics; both static and dynamic properties are covered. The compositional verification method imposes structure on the verification process. Because of the possibility of focusing at one level of abstraction (information and process hiding), compositional verification provides transparency and limits the complexity per level. Since verification proofs are structured in a compositional manner, they can be reused in the event of reuse of models or modification of an existing system. The method is illustrated for a generic model for diagnostic reasoning.     

Verifying Contextual Refinement with Ownership Transfer

Contextual refinement is a compositional approach to compositional verification of concurrent objects.There has been much work designing program logics to prove the contextual refinement between the object implementation and its abstract specification.However,these program logics for contextual refinement verification cannot support objects with resource ownership transfer,which is a common pattern in many concurrent objects,such as the memory management module in OS kernels,which transfers the allocated memory block between the object and clients.In this paper,we propose a new approach to give abstract and implementation independent specifications to concurrent objects with ownership transfer.We also design a program logic to verify contextual refinement of concurrent objects w.r.t.their abstract specifications.We have successfully applied our logic to verifying an implementation of the memory management module,where the implementation is an appropriately simplified version of the original version from a real-world preemptive OS kernel.     

基于信息修正的深度残差学习

提出了一种新的深度残差网络的拓展模块,有效提高了学习表示的鲁棒性。所提出的方法是一个简单的即插即用模块,即组卷积式编码-解码结构,它可以作为一个额外的信息过滤部件集成到原来的深度残差网络中。利用编码器的下采样来产生信息压缩过的特征图,解码器模块被驱动以产生激活准确的特征图,其能够突出显示输入图片中最具有判别力的区域,最后通过元素级相加和激活操作对输入特征进行信息修正。为了使设计的模型计算更加高效,通过减少残差分支的通道数来探究其轻量级版本的表现,发现并没有明显的性能下降现象。在各种基于残差网络的架构上进行实验,获得了一致性的性能提高,而且付出的计算代价与原始版本相比差别不大,甚至还低。     

Interactive tool support for CSP || B consistency checking

CSP || B is an integration of two well known formal notations: CSP and B. It provides a method for modelling systems with both complex state (described in B machines) and control flow (described as CSP processes). Consistency checking within this approach verifies that a controller process never calls a B operation outside its precondition. Otherwise the behaviour of the operation cannot be predicted. In previous work, this check was carried out by manually decomposing the model before preprocessing the CSP processes to perform a hand-written weakest precondition proof. In this paper, a framework is described that mechanises consistency checking in a theorem prover and removes the need for preprocessing. This work is based on an existing PVS embedding of the CSP traces model, but it is extended by introducing a notion of state so that the interaction between processes and machines can be analysed. Numerous rules have been defined (and proved) which enable consistency checking and decomposition via PVS proof. These rules also formally justify the relaxation of previous constraints on CSP || B architectures, thereby widening the scope of CSP || B modelling. The PVS embedding and rules presented in this paper are not only applicable to CSP || B specifications, but to other combined approaches which use a non-blocking semantics for the state-based operations. R. Lazic, R. Nagarajan and J. C. P. Woodcock     

基于Petri网的CSP并发系统验证技术研究

通信顺序进程(CSP)和Petri网是两种重要的并发系统建模工具。CSP语言具有高度抽象性,可有效刻画并发进程之间的各种相互作用,但在物理结构的描述与验证分析方面显得不足。Petri 网是一种形式化、图形化的并发系统建模和分析工具,侧重于系统的物理结构描述和性质分析。结合两者优点,首先利用CSP描述待验证的并发系统,然后将其转化为Petri网来分析系统的动态行为特性,最后利用性质分析工具TINA对系统性质进行分析和验证。实验结果表明,传统的CSP进程性质验证工具不能验证CSP进程的安全性,但其转化为Petri网后可有效地分析出导致安全性不能满足的危险因素,从而扩大了CSP描述的并发系统可验证性质的范围。     

标记迁移系统的组合可达性分析

标记迁移系统是一种在计算机辅助设计和验证中得到广泛使用的形式模型。当系统中的模块比较多时,系统的整体模型有可能出现状态空间的指数级爆炸,组合可达性分析是缓解这一问题的一种有效方法。已有的工作缺乏对该方法基本原理的清晰描述和精确表达。本文对其基本原理进行了分析和概括,并作了形式化陈述,证明了相关结论。本文的工作有助于深入理解和澄清组合可达性分析的内部工作机制。     

Specification, Refinement and Verification of Concurrent Systems—An Integration of Object-Z and CSP

This paper presents a method of formally specifying, refining and verifying concurrent systems which uses the object-oriented state-based specification language Object-Z together with the process algebra CSP. Object-Z provides a convenient way of modelling complex data structures needed to define the component processes of such systems, and CSP enables the concise specification of process interactions. The basis of the integration is a semantics of Object-Z classes identical to that of CSP processes. This allows classes specified in Object-Z to be used directly within the CSP part of the specification.In addition to specification, we also discuss refinement and verification in this model. The common semantic basis enables a unified method of refinement to be used, based upon CSP refinement. To enable state-based techniques to be used for the Object-Z components of a specification we develop state-based refinement relations which are sound and complete with respect to CSP refinement. In addition, a verification method for static and dynamic properties is presented. The method allows us to verify properties of the CSP system specification in terms of its component Object-Z classes by using the laws of the CSP operators together with the logic for Object-Z.     

面向源代码的软件模型检测及其实现

模型检测应用于检测软件可靠性具有重要意义.介绍了一种基于谓词抽象和反例引导抽象求精技术对源程序进行建模和验证的模型检测方法,并结合自行研发的Jchecker工具详细介绍了该软件模型检测技术的运作过程和关键算法.