Basically, I think recent years due to advancement of nanometer technologies,
analog design has been changed from analog alone to mixed mode design. So,
the best analog research should better necessarily benefit from this device shrunk, otherwise, the analog research is not very useful since it cannot come up with the digital pace.
I personally would think those require large digital part in the past and could not have done in the past should benefit the most now. For example, delta sigma type analog design like delta sigma ADC/DAC, frequency synthesizer, and DSP-based or calibration-based analog design since in the past, those requires a lot area and now the area is not a concern, so we should focus how we should make use digital logic to make analog even more precise than in the past, instead of just researching only analog parts without benefiting the present digital advancement