| 摘要 |
<p>A data coding method produces codewords with a scheme that changes for different codewords, and decodes codewords with a scheme that remains constant for all codewords. The coding method receives k user bits, codes the user bits to produce k + r output bits, corrupts any one of the output bits, and accurately reproduces at least k-r-1 user bits. The data coding method may code an input sequence (b0, b1, b2, ... , bk-1) to produce a codeword c = (c(-r), c(-r+1), ... , c(-1), c0, c1, ... , c(k-1)), where for 0 <= I <= k-1, symbols are determined according to the following Formula (II): c(i)f0 = c(i) = b(i) ⊕ c(i-1)f1⊕ c(i-2)f2⊕ ... ⊕ c(i-r)fr, where ci and bi are corresponding user data and codeword bits at time i. In this case, for each of a plurality of input sequences, appropriate initial conditions (c(-r), ..., c(-2), c(-1)) = s(b) are independently selecting to satisfy a predetermined property in the codeword coded according to Formula (II). Codewords coded using the appropriate initial conditions are output. For each codeword, the appropriate initial conditions are appended to the codeword coded therefrom.</p> |
