I'm working on a project which aims to make an MP3 file less fuzzy by adding virtual details, which are generated by an algorithm, to the file. I come up with an ide. But it seems that the idea is similar to the Fullsound, a technique developed by Philips, which I know roughly about how it works. Should I continue to work on it? Or are there any possible ways to make MP3 files less fuzzy...