how to simulate the crystal under the EM field?

Now, I have a MZ modulator used as a electric field sensor. That means when the MZ modulator is in the EM field, the laster throug the MZ modulator will be modulated and the output power will change with the amplitude of the EM field.
Therefore, I want to design the waveguide of the MZ modulator. If I want to use the FDTD method to simulate this structure, I need two sources. One is laster, another is EM field. Is there any FDTD software which can do this simualtion?
Thank you vey much.

For a field sensor, you will probably just assume the external field you are measuring is static. Even if that is at a microwave frequency, that is orders of magnitude slower than an optical frequency.

you mean, the external field can be applied in the model as same as the optical signal?

The most basic FDTD neglects static fields as linearity allows us to analyze these field separately. Without knowing anything about your problem, my thinking is that you will not have to incorporate the static fields into your model. Let me make a bunch of assumptions to illustrate what I mean:

I anm envisioning that you are placing some sort of electro-optic material (EO) or magnetorestrictive (MR) material into/onto one arm of your MZ sensor. That material alters its refractive index and/or dimension in response to a field. This affects the phase of the propagating light in that arm of the MZ which will then interfere with the reference arm to produce your measurement signal. You will first have to choose the value of the static field you are modeling. Second, you will need to understand how the EO/MR will respond to that field. After you know that, I see you just incorporating those material properties into your FDTD model. This way, your model does not have to directly account for the static field during the actual simulation.

I am curious why you are using FDTD to model an MZ interferometer. Can you just use an analytical model here? Are you not interested in sensing static fields, but more interested in the time response of your sensor? I can't help thinking there is a more efficient method you could be using. Again, I know so little about what you are doing it is difficult to have any confidence in things I may suggest. With that said, have you considered any frequency-domain methods like beam propagation method (BPM), method of lines (MoL), or finite-difference frequency-domain (FDFD)? If you give me some more information about your problem, I may be able to steer you down a more efficient path.

-Tip