Disclosed is an error correction in reproducing a digital signal of a digital audio system or a PCM reproducing system.
Code words such as doubly-encoded Reed-Solomon code comprise first code blocks including a plurality of information words arranged in a first arrangement and a plurality of first check words generated by a code having a Hamming distance of d, to said information words, and second code blocks including a plurality of information words arranged in a second arrangement and a plurality of first check words each having a plurality of information words and a plurality of first check words belonging to different one of said first blocks, and a plurality of second check words generated by a code having a Hamming distance of d 2 to said plurality of information words and said plurality of first check words, are decoded to correct errors. For such a doubly-encoded error correction code, at least error detection is effected in a first decoding and flags indicating decoding conditions are added. In a second decoding, error detection for the first code blocks and error correction for S 2 words at unknown locations and E flagged word erasure, where S 2 and E satisfy a relation of 2S 2 + E S d, - 1, are parallelly or sequestially are effected, and a combination of S 2 and E having a high correction capability and a low probability of mis-correction is selected from a plurality of correction results of error locations and the number of flags added at the first decoding, and the word errors are corrected based on the selected combination.