Stopping criteria for iterative decoding based on mutual information

In this paper we investigate stopping criteria for iterative decoding from a mutual information perspective. We introduce new iteration stopping rules based on an approximation of the mutual information between encoded bits and decoder soft output. The first type stopping rule sets a threshold value directly on the approximated mutual information for terminating decoding. The threshold can be adjusted according to the expected bit error rate. The second one adopts a strategy similar to that of the well known cross-entropy stopping rule by applying a fixed threshold on the ratio of a simple metric obtained after each iteration over that of the first iteration. Compared with several well known stopping rules, the new methods achieve higher efficiency.