Decoding method and apparatus

Abstract

A method and apparatus for decoding forward error correction (FEC) encoded data. A stream of units of FEC encoded bits are received, where the units are derived from a transmitted signal, where each unit represents a one-bit data value, and where each unit includes correctness bits that together reflect a level of probability that the one-bit data value represented by the unit differs from corresponding bits in the signal before it was transmitted. The stream of units of FEC encoded bits are decoded by using the quality level of bits to perform soft-decision convolution decoding on the stream of units of FEC bits, where the soft-decision convolution decoding produces, for block decoding, a stream of symbols made up of bits. In the stream of units of FEC bits, subsequences of units that are prone to erroneous soft-decision convolution decoding are detected by determining, for the sub-sequences whether the distribution of quality bits indicate the units are below a threshold level of correctness, and by comparing characteristics of that distribution to a given set of characteristics predetermined to be prone to result in incorrect decoding. Among the symbols produced for block decoding, symbols containing one or more bits that correspond to units that were in subsequences detected to be prone to erroneous soft-decision convolution decoding are flagged or tagged, and thereby, errors in the stream of symbols may be soft-decision block decoded by erasing the flagged symbols.

1 . A method of predicting when an error correction module will fail to correct errors contained in a quality ranked data stream, comprising: 
 determining an acceptable concentration of bits in the data stream that have a quality below a quality threshold; and    predicting that the error correction module will fail to correct errors when the concentration is exceeded.    
   
   
       2 . The method according to  claim 1 , wherein the concentration is expressed as M out of N, where N is a number of bits comprising a sequence of bits in the data stream, and M is a count of bits, in the sequence, that have a quality below the quality threshold.  
   
   
       3 . A method of decoding, comprising: 
 generating a first-decoded stream of bits by decoding a stream of forward error correction (FEC) encoded bits based on a level of correctness of each of the FEC encoded bits; and    decoding the first-decoded stream based on the level of correctness.    
   
   
       4 . The method according to  claim 3 , wherein the level of correctness is based on a quality of a signal from which FEC encoded bits are derived.  
   
   
       5 . A method of decoding, comprising: 
 identifying bits among first bits, according to correctness levels of the first bits;    decoding the first bits into second bits; and    decoding the second bits in accordance with the identified bits of the first bits.    
   
   
       6 . The method according to  claim 5 , wherein the correctness levels are based on a quality of a signal from which FEC encoded bits are derived.  
   
   
       7 . A method of decoding, comprising: 
 decoding first bits into second bits;    identifying bits, among the first bits, having a probability of erroneous decoding by said first decoding, said identifying being based on correctness levels of the first bits, and being performed separate from said decoding of the first bits;    flagging bits of the second bits that correspond to identified bits among the first bits; and    decoding the second bits after said flagging.    
   
   
       8 . The method according to  claim 7 , wherein the levels of correctness are based on a quality of a signal from which the first bits are derived.  
   
   
       9 . A method of FEC decoding, comprising: 
 identifying sequences of FEC encoded first bits, the sequences having a number of bits that have a low intrinsic quality level, and the sequences being identified based on such number;    Viterbi decoding the FEC encoded first bits and the identified sequences into second bits; and    erasure block decoding the second bits by erasing certain bits of the second bits, or symbols that contain the certain bits, where the certain bits correspond to bits in the identified sequences of first bits.    
   
   
       10 . The method according to  claim 9 , wherein the number of low intrinsic quality level bits in the sequences is determined by counting a number of low intrinsic quality level bits within a sliding window of the FEC encoded first bits, the sliding window preceding said Viterbi decoding, and wherein the intrinsic quality level of the bits is based on a quality of a received transmission signal from which the bits are derived.  
   
   
       11 . A method of decoding forward error correction (FEC) encoded data, comprising: 
 receiving a stream of units of FEC encoded bits that are derived from a transmitted signal, where each unit represents a one-bit data value, and each unit comprises correctness bits together reflecting a level of probability that the one-bit data value represented by the unit differs from a corresponding one-bit value in the signal before it was transmitted;    correcting errors in the stream of units of FEC encoded bits by using the correctness bits to perform soft-decision convolution decoding on the stream of units of FEC bits, where the soft-decision convolution decoding produces a stream of symbols for block decoding, the symbols comprising bits;    identifying, in the stream of units of FEC bits, subsequences of units that are prone to erroneous soft-decision convolution decoding, by determining, for the sub-sequences, a total number of units whose correctness bits indicate the units are below a threshold level of correctness, and by comparing the total number to a given number that is based on or equal to a total number of units in the subsequences;    flagging for erasure symbols among the produced symbols, the flagged symbols comprising one or more bits that correspond to units that were in subsequences identified to be prone to erroneous soft-decision convolution decoding; and    correcting errors in the stream of symbols by block decoding codewords comprised of symbols including symbols not tagged for erasure.    
   
   
       12 . The method according to  claim 11 , wherein the soft-decision convolution decoding comprises soft-decision Viterbi decoding, and wherein the block decoding comprises Reed Solomon decoding.  
   
   
       13 . A decoding apparatus, comprising: 
 a detector detecting a portion of data with a probability of being erroneously decoded by a convolutional decoder;    a convolutional decoder convolutionally decoding the data; and    a block decoder further decoding the data by taking into account that the data has a portion that has been detected to have a probability of being erroneously decoding by said convolutional decoder;    wherein the further decoding of said block decoder is performed by block decoding with erasure of data corresponding to the detected portions.    
   
   
       14 . A method of decoding, comprising: 
 identifying portions of a data stream that are likely to be erroneously convolutionally decoded based on a quality level of the data stream;    wherein the data stream containing the portions is convolutionally decoded.