What would you suggest? MWS or HFSS for this case?

I have been trying to find the far field radiation pattern of a monopole antenna placed inside a photonic crystal using MWS. Transmission and S11 simulation results match the experimental results. But somehow far field radiation pattern isn't even close to experimental data. I ve run the simulation multiple times with many possible settings. Still no sensible result. I ve suspected that the reason for inconsistency is related to the high Q value near the band edge frequency. Since the group velocity near the band edge is very low these frequencies decay very slowly. Would it be better to use HFSS for this structure? I would appreciate your suggestions. I may add good results to a publication if I can get some. I have attached the structure (.mod file and the .sat file are included in the .rar file).

Hi irfan1,
I guess it would be pretty much hopeless to simulate this structure in HFSS. The structure is extremly large.
At 11 GHz and taking into account the high dielectric constant of the material you will have a box of 20*20*20 wavelength. Using just the basic discretisation for a second order methode (lambda/4) you will have 80*80*80 tets-> more the 500 000 tests just for the basic mesh. This does not include any geometry refinement for the coax or slabs.
I don't think HFSS can handle this ....
Any comments from our HFSS experts concerning this problem?
F.

Dear RFSimulator,
I guess you are very knowledgable on MWS would you have any suggestions specific to my structure. Suggestions that can improve the simulation. Is anything set wrong in the simulation setup, such as the distance beetween the boundary and structure or can I improve the mesh in some way? Would you suggest the use of any symmetry condition for this structure.
Thanks in advance

Hi irfan1 ,
I will run your simulation as soon as I have some sparetime available to see what I can do.
F.