Coding approaches to fault tolerance in combinational and dynamic systems (Q2770033)

A fault-tolerant system is able to tolerate internal faults and preserve desirable overall behavior and output. A necessary condition for a system to be fault-tolerant is that it exhibit redundancy, which enables it to distinguish between correct and incorrect results or between valid and invalid states. This book consists of two parts. NEWLINENEWLINENEWLINEThe first part deals with fault-tolerant combinational ( i.e., memoryless) systems. Chapter 2 reviews von Neumann's work on ``Probabilistic Logics and the Synthesis of Reliable Organisms from Unreliable Components'', the first systematic approaches to fault-tolerance, including some more recent work on noisy gates. Work on arithmetic coding and algorithm-based fault-tolerant scheme in algebraic settings is described in Chapter 3. NEWLINENEWLINENEWLINEThe second part of this book focuses on fault-tolerance in dynamic systems, such as finite-state machines. This part describes a systematic methodology for adding structured redundancy to a dynamic system, exposing a wide range of possibilities between no redundancy and full replication. These possibilities can be parameterized in various settings, including algebraic machines (Chapter 4) and linear dynamic systems (Chapters 5 and 6). NEWLINENEWLINENEWLINEChapter 7 relaxes the traditional assumption that the error-correcting mechanism does not fail by using distributed error-correcting mechanism. The final chapter presents a methodology for fault diagnosis in discrete event systems that are described by Petri nets; coding techniques are used to detect and identify failures. NEWLINENEWLINENEWLINEThis book emphasizes resource-efficient methodology that avoid excessive use of redundancy by exploiting the algorithmic/dynamic structure of a particular combinational or dynamic system.