Which 3D EM Simulator for antennas (UWB, antenna array, nanostructures)

Hello,

I'm searching for "the best tool" to design ultra-wideband antennas and antenna arrays with a highly directional response pattern.
Possible geometries are e.g. bow-tie, vivaldi or horn antennas.

The frequency range to be simulated is DC to 10 GHz.

Later I also want to analyze nanostructures like nanowires or carbon nanotubes and their behavior if formed like an antenna. Since these structures appear to be electrically very large (e.g. 2cm in length, 100nm diameter), I guess I need a

Of course I'm also interested in simulating the response of a scene (e.g. a certain material or a person) if a signal is radiated by one antenna and detected by either the same or another antenna. This may also take multireflections into account.

Basically I want to simulate the 3D radiation pattern, the S11 parameter and the gain of the antennas. Furthermore I want to simulate the signal being reflected by a scene and being detected by another antenna.

The software tool I already though about are
- Ansys HFSS
- Agilent ADS
- CST MWS
- Comsol RF module
- FEKO
- WIPL-D Microwave
- AWR AXIEM


Is there any other software, which I could use for my application?
What is your experience with these tools in a similar field of application?

Regards,
Daniel

hi daniel
i think the best software for simulation is CST but drawing in CST hard but HFSS good for some UWB application and for HFSS is week and for nano scale i never seen paper in HFSS or CST simulation and i think for this structure better simulate with numerical method programing or may be XFDTD software useful

Hi Daniel

I am working on similar topic as you. I study the properties of nanowires at high frequency. So far, i have tried ads and cst. These two software gives me similar S-parameters. Ads is easy to start with. With cst, as i tried to scale down, the simulation became really long and i encountered memory exeption error. i think it has something to do with the mesh size. just for your information.

HSU

i recommend CST for all simulation kinds

A time-domain code would probably be best for something that broad. MWS and XFDTD are pretty obvious choices there. XFDTD has a very fast GPU implementation if you have a CUDA-capable card. That could help get results quickly.

I would also recommend some time domain code for the antenna array. Empire XCcell (based on FDTD) is a good choice for antennas.
It is very fast (>2000 Mcells/s) without accelerator card, and has cluster capability for very big models.

EMPIRE XCcel Antenna Examples
EMPIRE XCcel - FDTD Simulator for 3D EM Modeling