Is there some kind of noise-tolerant algorithm that ignores errors in the incoming buffer, without storing the original state?

Question

Hashing

Interested in hashing that is not sensitive to small changes in the initial data with a mutation rate of up to 1%.
That is, with any modification, with any distribution of 10 bytes out of 1024 information, the output has always been the same buffer of at least 32 bytes. But when modifying say 20 bytes, the output buffer was different already.

Answer 1

Will Hamming code work?