Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Optimizing 3D structures in IE3D

Status
Not open for further replies.

mengjom

Newbie
Joined
Jan 29, 2007
Messages
2
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,307
optimization with ie3d

I am currently working on simulating a so-called ESPAR antenna in IE3D. The antenna consists of one active element (monopoloe) in the centre of a limited groundplane and 6 parasitic elements (also monopoles) placed in a circle around the active element (in the simulations the parasitic elements are loaded with lumped circuits to adjust the antenna radiation pattern).

My question relates to the optimization of the antenna. Initially I have two optimization parameters. One is the distance from the active element to the parasites (i.e. the radius of the circle which the parasites are placed on), and the other is the element height (of all the 7 elements).
I have been sucessful in defining the radius as an optimization variable, but so far I have not been able to define the element height as a variable. What I essentially want to do is to define the Z-coordinate of a layer (the layer in which all the vertices of "the top" of each antenna element lie in) as an optimization variable.

Maybe it is a lot more straightforward than I think, but all help is appreciated!
 

magnetic field pattern ie3d

Hi, Mengjum:

On the polygon based layout editor MGRID, you can only define the xy coordinates of a group of vertices as one optimization variable. You can also assocate the xy coordinates of multiple group of vertices with one variable. However, you can define the z-coordinate as an optimization variable on MGRID. Once we were thinking about to expand this capability. However, we decided to implement it onto IE3DLIBRARY, the object oriented layout editor of IE3D.

On IE3DLIBRARY, you can optimize very complicated 3D structures. We have documented an example for a wire bond and two coupled wire bonds. It is in the Zelaand Virtual Training in www.zeland.com. In the examples, we allow users to define the size and the height of a wire bond, and the separation of 2 wire bonds as optim variables. You can even do a real-time EM tuning on them. You can slide the bars on the FastEM Design Kit of IE3D V12 and see how the variables are changing the shapes of the geometry and the results simultaneously.

Regards.
 

ie3d zeland software latest versions

In 1993, Zeland Software, Inc. introduced the IE3D Electromagnetic Simulation and Optimization Package. As a method of moment (MOM) simulator, IE3D has many good features in modeling planar and 3D circuits and antennas in layered dielectric environment. The IE3D is easy-to-use, accurate and efficient in modeling wide range structures such as microstrip circuits and antennas, strip-line circuits, CPW circuits and antennas, coaxial structure with uniform dielectric filling, inverted-F antennas, dipoles and other wire antennas, high speed transmission lines, high speed digital circuit interconnects, high speed digital circuit packaging. However, moment method codes have some inherent disadvantage in modeling 3D dielectric structures, waveguide structures and structures emphasizing near field distribution. For this reason, we introduced the FIDELITY Electromagnetic Simulator in November 1997.
FIDELITY is a finite-difference time domain (FDTD) based full-wave electromagnetic simulator. In the last 20 years, much research has been focused on the development of FDTD algorithms. Compared to frequency domain simulation algorithms such as MOM and finite element method (FEM), FDTD has the following unique features:

1.FDTD is easy to implement. Its basic principle is to use finite difference to represent the differentials in the Maxwell’s equations. By employing the Yee-algorithm, we combine the electric field and magnetic field together to convert the Maxwell’s equations into algebraic equations.

2.The final algebraic equations for FDTD are in the time-marching style. FDTD does not create large matrix equations which are common in MOM and FEM. The memory requirement for FDTD is proportional to N compared to N to N2 for FEM and N log(N) to N2 for MOM. The time requirement of FDTD is also more proportional to N compared to N2 for FEM and N2 to N3 for MOM. Although the basic computational requirement for FDTD is normally much higher than that for MOM for modeling small and medium size structures, for large structures, FDTD may require much less computational resources than MOM. On the other hand, FDTD normally requires much less computational resources than FEM.

3.A FDTD simulation normally can yield a wide-band frequency response. MOM and FEM normally require sweeping in frequency domain.

4.FDTD simulators can handle complicated dielectric structures much easier than MOM and FEM.

As it is, the FDTD simulator FIDELITY is complementary to the MOM simulator IE3D developed by Zeland Software, Inc.

The basic algorithm on FDTD can be found in numerous textbooks and research papers. We will not repeat it here. In this chapter, we will mainly focus on the configuration of the FIDELITY software package.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top