next up previous
Next: C. METHOD 2 Up: B. METHOD 1 Previous: Huffman Coding

How to decode

The information in the compressed file is sufficient to reconstruct the original file. This is done with the ``uncompress'' algorithm, which works as follows: again, for each character $c^f_m~(m>n)$, the predictor (sequentially) emits its output $ P^f_m$ based on the $n$ previous characters, where the $c^f_l$ with $n < l < m$ were obtained sequentially by feeding the approximations $P^f_l(k)$ of the probabilities $ Pr(c^f_l = z_k \mid c^f_{l-n}, \ldots, c^f_{l-1})$ into the inverse Huffman Coding procedure. The latter is able to correctly decode $c^f_l$ from $code(c^f_l)$. Note that to correctly decode some character, we first need to decode all previous characters.

Juergen Schmidhuber 2003-02-19