Simulation and Measurement of WLAN Antenna

Designs using DOS and Windows

author: Trevor Marshall

Click for Trevor's Home Page


Link to my tutorial 'Antennas Enhance WLAN Security' in, October, 2001

Starting Out - Installing the software

NEC2 software was written in Fortran. Although some commercial versions can be purchased which run directly with the Windows GUI, the freely available versions operate in a DOS command-line drive environment.

You can download the DOS executables from this link. I use the 4NEC2 GUI program written by Arie. It takes the raw data in the NEC2 output files and displays it graphically in a visually usefull form. All the diagrams in my articles (and on the home page) were created with 4NEC2. You can also download 4nec2, free of charge, from "The (unofficial) NEC Archives". The new 3-D solids-renderer is superb....

4NEC2 makes the NEC2 executables run automagically in a DOS window of Windows98, 2000 or XP. NEC2D analyses antennas by dividing them up into many small segments, and calculating the current and properties of each segment. I found that Windows 2000 in a computer with 256K of DRAM can run a 4000 segment analysis, but it takes about 5 minutes to swap all its other programs out of core. You see, the 4K model uses almost all the 256K, and NEC's DRAM has to be contiguous (cannot be swapped by virtual memory). My slim Windows 98 OS also ran executables up to 4000 elements. These are very complex models, however, and the 1k9 executable managed to run all of my complex models, even those with multiple reflectors.

Orientation issues

4NEC2's GUI assumes that the antenna is located along the Y axis. Normally, NEC models assume propogation along the Z axis. You will get used to using the GM card to rotate your models around the axes. It is the easiest way that I found to obtain both azimuth and elevation plots.

NEC2 is written for input data files organized as "card decks". No comment cards can be used (except for the very front at the file) and it is sometimes quite hard to work around the limitations of the input format. Those of us who once had to use card punch machines, however, will welcome the 'free-format-input', which does not force the input data to be placed in certain column positions on the 'card'.

4NEC2 has an edit menu item to edit the NEC input file. You can then compile the files while running NEC2D in a DOS window (Start:Run under 2000 and XP) or from 4NEC2 itself. Finally review the results with the 4NEC2 GUI.

The BiQuad antenna

On my home page is my BiQuad design for a parabolic dish feed. The BiQuad also makes a useful stand-alone antenna, with wide bandwidth and about 11dBi gain (they work well to boost PCS cell-phone reception). Simple to make, and with good performance, the BiQuad antenna has become a workhorse around my laboratory! I also find that its simple construction makes it almost foolproof to build, and have even used BiQuads as a reference gain antenna from time to time...

Prior to NEC2 analysis, the reflector is broken up into a mesh of 0.05 or 0.1 wavelength segments. This can be done with just a few cards, using the GM command. A copy of the NEC2 sourcefile can be found at this link, and the corresponding output is at this link. An edited set of graphics from 4NEC2 is shown below.

WLAN antenna simulation with NEC2 will save you countless hours of cut and try in the lab. Another model that you might simulate is the ubiquitous Coffee Can feed. Lower gain than the BiQuad, it also tends to radiate significant sidelobes. I produced an optimum design for a Coffee Can Antenna, with about 9dB of gain. You can copy the NEC model from this link.

WLAN BiQuad with an N type connector
Biquad visualized with 4NEC2

Link to article 'Antennas Enhance WLAN Security' in, October, 2001

Click here for a list of other articles written for the E.E.Times CMP Publishing Group

Click here for a general list of publications on Analog, Digital and Software Design

DISCLAIMER: Any resemblance between the above views and those of my employer(s) are purely coincidental. Any resemblance between the above and my own views is non-deterministic. My existence can be challenged. The question of the existence of views in the absence of anyone to hold them is left as an exercise for the reader. The question of the existence of the reader is left as an exercise in the second order coefficient.

All information published at this website is (C)Copyright 1995-2001 Trevor Marshall.
All rights are expressly reserved. Click to email webmaster