We consider the problem of broadcasting a message from one processor (
called the source) to n other processors, We adopt the 1-port half-dup
lex model in which every processor (node) can communicate with at most
one other node in a unit of time and during this period one of the co
mmunicating nodes can only send information and the other can only rec
eive it. The source is fault-free but all other nodes are subject to p
ermanent failures. A faulty node can receive the message but cannot re
lay it. The fraction of faulty nodes is bounded by a constant. We cons
ider nonadaptive broadcasting algorithms working correctly in the pres
ence of faulty nodes and investigate their worst-case running time, We
also show lower bounds for broadcasting time under this scenario. Our
main result is the construction of a fault-tolerant broadcasting algo
rithm whose running time is less than 1.73 times larger than optimal,
for sufficiently large n. (C) 1997 Academic Press.