I had similar experience before when I started transiting my research work from 3G to 4G and beyond. It seems that every a few years, researcher and students ask similar questions. Here is what I was pondering over before and hope my then thoughts and my methodology would help you a little bit, though it won't give you specific research topics right way.
[What Is The Next for Mobile System Design? I: A Single-Cell Model Perspective on Downlinks]
As you may see, what I did may be described in three steps essentially:
to compare what we have achieved with then what people want from channel models perspective.
Afterwards I basically examined what kind technologies might possibly help push what we have achieved towards now what people want from a channel model or system model perspective.
pick the one you are interested in and which can provide largest realistic performance improvement
In my then analysis and then examples, however, I only used then-popular and standardized 3GPP COST-231 single-cell SISO channel model. Depending on your needs or interest, if you want to reuse my methodology above, I would strongly suggest you replacing my analysis with your own channel model or system model or the ones you choose.
Please let me know if you have any other thoughts or thing.