New methods for evaluating the impact of single event transients in VDSM ICs

Abstract

ISBN: 0769518311This work considers a SET (single event transient) fault simulation technique to evaluate the probability that a transient pulse, born in the combinational logic, may be latched in a storage cell. Fault injection procedures and a fast fault simulation algorithm for transient faults were implemented around an event driven simulator. A statistical analysis was implemented to organize data sampled from simulations. The benchmarks show that the proposed algorithm is capable of injecting and simulating a large number of transient faults in complex designs. Also specific optimizations have been carried out, thus greatly reducing the simulation time compared to a sequential fault simulation approach

Design of a soft-error robust microprocessor

“…Some works dedicate special attention to analyze the probability of a SET becoming an indirect SEU (HASS et al, 1998;HASS, 1999;MASSENGILL et al, 2000;NICOLAIDIS, 2002). Other approaches like a SET propagation method based on topological timing analysis could be used to evaluate such probability too.…”

“…Evaluation resources like (MASSENGILL et al, 2000;LIMA et al, 2001-a;NICOLAIDIS, 2002; might offer some answers concerning the effectiveness of mitigation techniques against indirect and direct SEUs on the circuits. However, these works and other many ideas about this issue even lack advances to become in consolidated and practical analysis tools for large circuits.…”

Designing fault-tolerant techniques for SRAM-based FPGAs

“…5 When a charged particle strikes a memory cell's sensitive nodes, such as a drain in an off-state transistor, it generates a transient current pulse that can mistakenly turn on the opposite…”

An accurate probabilistic model for error detection

“…Probabilities are inferred by a complete set of samples or instantiations that are generated for each node in the network according to local conditional probabilities stored at each node. The advantages of this inference are that: (1) its complexity scales linearly with network size, (2) it is an any-time algorithm, providing adequate accuracytime trade-off, and (3) the samples are not based on inputs and the approach is input pattern insensitive. The salient aspects of the algorithm are as follows.…”

“…In this work, we focus on the probabilistic modeling of SET that allows an SET generated at a node captured by a latch and measure it by the fault detection probability of that node. Existing estimation techniques for SET [2,3] rely on simulation and hence are modeling the pattern dependence of SET by estimation methods that are in itself pattern-sensitive. We use a probabilistic zero-error model for detection probability of errors that can be uniformly applied to permanent stuck-at faults as well as soft transient errors predominant in nano-domain.…”