Correcting Deletions in Probabilistic Non-Binary Segmented Burst Deletion Channels

Abstract

Consider a burst deletion channel where in a block of L consecutive non-binary symbols at most a single burst deletion of length b symbols exists. Existing schemes for this scenario leverage non-binary de Bruijn sequences to perfectly locate deletions. In contrast, we propose to solely use binary marker patterns in combination with a new soft-decision decoder. In this scheme, deletions are soft located by assigning a posteriori probabilities for the location of every burst deletion event of length of at most b, and replaced by erasures, then the resulting errors are further corrected by an outer channel code. Such a scheme has an advantage over the non-binary scheme as it in general increases the rate compared to an approach based on non-binary de Bruijn sequences, with only a minor loss in performance of locating a deletion error. Also, the proposed scheme provides a better error correction performance than existing schemes for the same code rate.