In theory, yes you can. However, I see two main problems with this:
1) A conference is less likely to accept a paper that has no actual data in it, i.e. you are basically just reviewing the literature
2) If you do get to present, someone might steal your ideas and do the experiments before you do!
To be honest, I would advise against doing this. You are basically giving your ideas away, and you still don't know if they would actually work. You would be better waiting until you have at least some new data to present, so that even if someone else takes your idea you will be a few steps ahead of them. Alternatively, you could write a review style paper (for publication in a journal) that would neatly show where the gap in the field lies, and then you could go ahead and conduct your experiments, and write another paper that neatly fills that gap.
This feels to me more like the sort of thing you would present internally at your uni, rather than at an external conference.