0
$\begingroup$

Can anyone think of a way to use Huffman Encoding to improve the compression rate of Lempel-Ziv Encoding?

My initial idea was to encode with higher probabilities given to shorter strings (i.e. the string "e" would have a higher probability than the string "ef").

Does anyone know of a good way?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
2
$\begingroup$

Huffman encoding is already used this way.

Huffman encoding is already used together with LZ77 to improve its compression rate. The most widely used instance of LZ77-style compression is DEFLATE. DEFLATE combines LZ77 with Huffman encoding. See https://en.wikipedia.org/wiki/LZ77_and_LZ78#Implementations and https://en.wikipedia.org/wiki/DEFLATE.

Huffman encoding is commonly used together with standard LZW (Lempel-Ziv-Welch) encoders to improve their compression rate. See the Wikipedia article on LZW and other standard resources on LZW coding for details: https://en.wikipedia.org/wiki/Lempel%E2%80%93Ziv%E2%80%93Welch

Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.