I would like to publish a journal paper and an ArXiV preprint, where I validate my results using the dataset available here (scroll down to the Files section). I would like to submit my paper to IEEE Transactions on Signal Processing (TSP). My question is, do I need any approval from the authors? I would, of course, cite the paper and the website. My thinking is that they voluntarily uploaded their data online. They didn't specifically state the data can be used in other publications, but they give a bibtex citation format right after the files section. I cannot find the relevant TSP guidelines on this, but if I could avoid contacting the authors (while being fully ethical) I would prefer it. What is your opinion about it? Thanks a lot!
It is common to use publicly available data. You need to cite the source of the data you used, but you do not need explicit permission from the authors of the data. As you said, they made it public.
(Caveat to this paragraph: I am not a lawyer). To be thorough, you might want to spend time poking around to see if there is a license that explicitly lays out what you can and cannot do with the data. However, I've never come across a situation where published data is off limits for use in a scientific study, the limitations are usually around things like selling the data or passing it off as your own.