Endre Selényi

Probabilistic system-level fault diagnostic algorithms for multiprocessors

Massively parallel computers (MFCs) introduce new requirements for system-level fault diagnosis, like handling a huge number of processing elements in a heterogeneous system. They also have specific attributes, such as regular topology and low local complexity. Traditional deterministic methods of system-level diagnosis did not consider these issues. This paper presents a new approach, called local information diagnosis that exploits the characteristics of massively parallel systems. The paper defines the diagnostic model, which is based on generalized test invalidation to handle inhomogeneity in multiprocessors. Five effective probabilistic diagnostic algorithms using the proposed method are also given, and their space and time complexity are estimated.

Dependability analysis in HW-SW codesign

The increasing complexity of todays computing systems necessitates new design methodologies. One of the most promising methods is hardware-software codesign, that supports unified hardware-software modeling at different levels of abstraction, and hardware-software synthesis. As applications include even critical applications, dependability becomes an important design issue. A novel approach for the underlying modeling in hardware-software codesign is presented in this paper. The basic idea of this new method is the extension of the descriptions of the functional elements with the models of fault effects and error propagation at each level of the hardware-software codesign hierarchy. From the extended system model various dependability measures can be extracted. This paper concerns test generation, solved by a generalized form of the well-known logic gate level test generation algorithms and extraction of the input model of integrated diagnostics, allowing testability and diagnosability analysis of the system.<<ETX>>

Modeling Uncertainty in System-Level Fault Diagnosis Using Process Graphs

This paper presents a novel approach to solving the probabilistic diagnosis problem in multiprocessor systems. The main idea of the algorithm is based on the reformulation of the error propagation model as a Process Network Synthesis (PNS) problem. This idea is illustrated by deriving a maximum likelihood decision procedure. The diagnostic accuracy of the solution is considered on the basis of simulation measurements, and a method of constructing a general framework for different aspects of a complex problem with the use of P-Graph models is demonstrated.

Quality control system for biomedical instrumentation

The authors describe the self-test and final test of a three-channel ECG device. It is also shown how the self-test routines are used for calibration, final test, and field service. In order to promote testing, an almost hardware-level controllability is assured through the remote-control connector. Though the instrument has the usual built-in calibration signal, an extra calibration signal is available which serves for testing the analog channels together with the electrode cables during service.<<ETX>>