Raymond T. Yeh

A prototyping language for real-time software

PSDL is a language for describing prototypes of real-time software systems. It is most useful for requirements analysis, feasibility studies, and the design of large embedded systems. PSDL has facilities for recording and enforcing timing constraints, and for modeling the control aspects of real-time systems using nonprocedural control constraints, operator abstractions, and data abstractions. The language has been designed for use with an associated prototyping methodology. PSDL prototypes are executable if supported by a software base containing reusable software components in an underlying programming language (e.g. Ada). >

FUZZY RELATIONS, FUZZY GRAPHS, AND THEIR APPLICATIONS TO CLUSTERING ANALYSIS*

Publisher Summary This chapter discusses the basic terminologies and notations regarding fuzzy relations and fuzzy graphs. The the concept of similarity relation introduced by L. A. Zadeh is derivable in much the same way as equivalence relation. Moreover, through this derivation, the resolution identity is brought out quite naturally. The chapter analyzes fuzzy graphs from the connectedness viewpoint and the presents the application of results to clustering analysis and modeling of information networks. The usual graph-theoretical approaches to clustering analysis involve first obtaining a threshold graph from a fuzzy graph and then applying various techniques to obtain clusters as maximal components under different connectivity considerations. These methods have a common weakness, namely, the weight of edges are not treated fairly because any weight greater (less) than the threshold is treated as 1(0).

Formal Specification and Verification of Distributed Systems

Computations of distributed systems are extremely difficult to specify and verify using traditional techniques because the systems are inherently concurrent, asynchronous, and nondeterministic. Furthermore, computing nodes in a distributed system may be highly independent of each other, and the entire system may lack an accurate global clock.

A commonsense management model

A description is given of a process-management model, called cosmos, which combines the best of existing models by incorporating three perspectives: activity, communication, and infrastructure. Cosmos is designed to manage a large software system from beginning to end. Current problems in system development are reviewed, and the principles on which cosmos is based are outlined. The reasons for using such a three-dimensional model are examined. The use of cosmos is described.<<ETX>>

Strong verification of programs

The authors investigate the strong verification of programs using the concept of predicate transformer introduced by Dijkstra (1974). They show that every do-while program has a loop invariant that is both necessary and sufficient proving strong verification. This loop invariant is shown to be the least fixpoint of a recursive function mapping predicates to predicates that is defined by the program and the postcondition.

Induction as the Basis for Program Verification

We will consider the inductive mechanisms in five techniques for verifying iterative/recursive program structures: inductive assertion, predicate transformers, subgoal induction, computation induction, and structural induction. We will discover that all five techniques can be justified by a single theorem about inductive proof techniques. We will also show that all five techniques face the problem of finding properties that will carry an induction. Such properties are called inductive sets. We will see that the inductive sets of the five techniques are easily related to one another and that a program proof by any of the techniques can be easily converted to a proof by any of the other techniques. Our conclusion is that computer programs simply are inductive definitions of the functions they compute. Induction is the only method by which they can be proved. The problems of induction are therefore unavoidable.

SEES&#8212;A Software testing Environment Support System

SEES is a database system to support program testing. The program database is automatically created during the compilation of the program by a compiler built using the YACC compiler-compiler.

Specification and verification of liveness properties of cyclic, concurrent processes

A technique is described for software specification and verification of concurrent, distributed systems. The complete specification of a program is given in terms of a hierarchical structure of module specifications. Module external specifications are abstract; module internal specifications are descriptions of internal implementations, either in terms of submodules or actual code. The verification that an implementation satisfies its specification is language independent for the former and language dependent for the latter. Distinguishing the liveness properites provided by a module and the liveness properties required by a module (from its comodules) allows the specification and verification of a given module to be independent from the specification and verification of its comodules.

A programming environment framework based on reusability

We propose a software life cycle which consists of two phases: rapid prototyping and automatic program generation. Software reusability holds the biggest leverage for both phases. This paper outlines the requirements for supporting this two-phase life-cycle using a software-base as an advanced software management system.

On periodicity of sequential machines

It was shown by Gill and Flexer [3] that the necessary and sufficient condition for the existence of nontrivial periodic decomposition of a sequential machine corresponds to the existence of a nontrivial cyclic partition. We have, in this paper, characterized the existence or nonexistence of cyclic partitions of machines under various connectedness conditions. Also, their theory is generalized here utilizing the concept of cyclic covers. As a consequence of this generalization, the open problem posed by Gill and Flexer [3] is solved. Upper bounds of periodicity of sequential machines are obtained using Sperners theorem [4].