I would prefer to keep the power confined. I would prefer to emit the power as a plane wave. I am open to suggestions.
Yes it is. I estimate about 2 trillion g's.
Acceleration caused by gravity, caused by something as massive as a black hole, causes the whole EM spectrum to frequency shift. In contrast, emitting a repeating frequency shift is going to give you an acceleration field that is, perhaps, -50dB (significantly weaker). I don't know how to calculate it's strength. Boosting power will help (I'm sure). A 1 to 2GHz sweep every millisecond might, optimistically speaking, produce a measurable change in a 100.000 gram weight of +/- 0.01 grams; my scale reads in grams, not newtons or pounds. It would be an acceleration field strength of 10^-4 g's. That would be a $$\frac{\Delta f}{\Delta t}=10^{12}$$.
Ideally, I would prefer to perform a frequency shift from 400 to 800THz every microsecond; that would give you $$\frac{\Delta f}{\Delta t}=4x10^{18}$$, 6 orders of magnitude higher. Perhaps that would boost the acceleration field by 6 orders of magnitude to a 100g's.
I didn't want to think about the hazards of stray acceleration fields. Maybe, someday, engineers and physicists will design emitters that will keep the acceleration field very strong locally. Maybe it will fall off by -20dB per meter. If we could ever create acceleration fields as strong as 10^15 g's, which could break the speed of light barrier (with the right geometry and masterful engineering techniques), the field would drop to 0.001g's after $$20log(\frac{10^{15}}{10^{-3}g})=360dBm$$, -20db/meter, about 18 meters. So don't stand too close.