We present an approach to detecting `babbling' faulty nodes on a bus by using a bus guardian to listen to previous messages and deduce a window during which future messages should appear. In general, one cannot detect all erroneous messages, but the messages that are incorrectly classified can be bounded, and therefore can be taken this into account when doing worst case response time analysis.

BibTex Entry

@inproceedings{Broster2001a,
 author = {I. Broster and A. Burns},
 booktitle = {Proceedings of the Work-In-Progress Session, 22nd IEEE Real-Time Systems Symposium, YCS 337},
 category = {scheduling},
 editor = {G. Fohler},
 organization = {IEEE},
 pages = {25-28},
 publisher = {Department of Computer Science, University of York},
 title = {The Babbling Idiot in Event-triggered Real-time Systems},
 year = {2001}
}