Abstract | ||
---|---|---|
In this paper the design of a FIR filter with self checking capabilities based on the residue checking is analyzed. Usually the set of residues used to check the consistency of the results of the FIR filter are based of theoretic considerations about the dynamic range available with a chosen set of residues, the arithmetic characteristics of the errors caused by a fault and on the characteristic of the filter implementation. This analysis is often difficult to perform and, to obtain an acceptable fault coverage the set of chosen residues is overestimated. obtained result a and therefore requires that Instead, in this paper we show how using an exhaustive fault injection campaigns allows to efficiently select the best set of residues. Experimental results coming from fault injection campaigns on a 16 taps FIR filter demonstrated that by observing the occurred errors and the detection modules corresponding to different residue has been possible to reduce the number of detection module, while paying a small reduction of the percentage of SEUs that can be detected. |
Year | DOI | Venue |
---|---|---|
2007 | 10.1109/DFT.2007.23 | DFT |
Keywords | Field | DocType |
filter implementation,detection module,best set,self checking fir filter,fault injection campaign,exhaustive fault injection campaign,fir filter,acceptable fault coverage,chosen residue,fault injection analysis,residue checking,different residue,fir filters,fault coverage,dynamic range | Stuck-at fault,Dynamic range,Fault coverage,Computer science,Electronic engineering,Real-time computing,Self checking,Finite impulse response,Fault injection | Conference |
ISSN | ISBN | Citations |
1550-5774 | 0-7695-2885-6 | 1 |
PageRank | References | Authors |
0.38 | 11 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Salvatore Pontarelli | 1 | 368 | 54.05 |
L. Sterpone | 2 | 182 | 22.70 |
G. C. Cardarilli | 3 | 71 | 10.34 |
Marco Re | 4 | 194 | 35.03 |
M. Sonza Reorda | 5 | 1099 | 114.76 |
A. Salsano | 6 | 90 | 13.37 |
M. Violante | 7 | 1 | 0.38 |