During the last years decoding algorithms that make not only use of so
ft quantized inputs but also deliver soft decision outputs have attrac
ted considerable attention because additional coding gains are obtaina
ble in concatenated systems. A prominent member of this class of algor
ithms is the Soft-Output Viterbi Algorithm. In this paper two architec
tures for high speed VLSI implementations of the Soft-Output Viterbi-A
lgorithm are proposed and area estimates are given for both architectu
res. The well known trade-off between computational complexity and sto
rage requirements is played to obtain new VLSI architectures with incr
eased implementation efficiency. Area savings of up to 40% in comparis
on to straightforward solutions are reported.