If you have SpectreRF you can perform a PSS/PAC analysis and use that to compute the coefficient of pulling. But this assumes a small interferring signal. You cannot use this approach to determine when injection locking occurs. For that you would have to use a transient analysis, which of course is an expensive simulation.

-Ken

If you use PAC, then the signal is injected at the 0th sideband, which is say at 1.8001GHz, then the lower sideband would actually be the -2 sideband at -1.7999GHz. It's got to be there because if you perturb an oscillator the response will primarily be in the phase, and you need both sidebands for phase modulation.

-Ken

Hi Ken, you mentioned that we can observe the injection locking occur. Could you advise how to do that with transient simulation. I design a injection locked VCO at 30 GHz with injection source at 10 GHz (third harmonic injection). Whatever frequency is injected (even at 5 GHz), the output frequency is always at 30 GHz (free-run oscillation frequency of the VCO). I run transient simulation and do dft to check the spectrum. Thanks.