Open Roads Radio forum for Ham, Amateur Radio and RV camping

Thread Tools Display Modes
Old 11-06-2017, 07:04 AM   #1
Junior Member
Join Date: Oct 2017
Posts: 7
Default How valuable is a system which can design antenna on its own

Hello everyone,

I am thinking of making a system which can generate a antenna for a given input radiation pattern, that is, we draw a radiation pattern we desire and the system churns out a antenna design which can achieve that? Is it trivial or is it valuable?
I have an idea of making a antenna design system but I don't know what input it should take. Any pointers?


I didn't find the right solution from the internet.

advertisement examples
JensenBreck is offline   Reply With Quote
Old 11-06-2017, 11:54 AM   #2
Senior Member
electricflyer's Avatar
Join Date: Jul 2013
Location: Douglasville, GA
Posts: 411

My son used to be a corporate training manager for a major cellular system service company and trained employees and contractors on antenna installation and the system itself. He said he had the hardest time getting engineers to understand that antenna design was theory, nothing set in stone. He and 200 others were laid off when the company said it lost most of its service contracts, except for the west coast. He was probably one of the most informed people in cellular service.
Marv KT4W
DW-Carolyn, 2 fur kids BooBoo and Gracie (felines)
Camped 71 days in 2016, Camped 33 in 2017, 29 booked for 2018
2014 Palomino Puma 25RL - TV 2011 F-150 SuperCrew 5.0
electricflyer is offline   Reply With Quote
Old 11-14-2017, 01:19 PM   #3
Carl, nn5i
NN5I's Avatar
Join Date: Sep 2011
Location: Tallahassee, FL
Posts: 1,399

Well, lessee. (Hunting for my system-designer hat and my project-manager hat -- haven't worn either one since decades before I retired. Oops, I'll need my consultant hat too.)

Parts of your system exist. In particular, NEC and the various packages that invoke NEC (such as MiniNEC, EasyNEC, and so forth) can accept an antenna description (details of antenna segments and feedpoints) and calculate a resulting pattern and other characteristics. NEC took many contributors several decades to develop.

What probably doesn't exist:

(1) an artificial language to describe mathematically the pattern you wish;

(2) a program that can translate the language (1) into the same format that NEC uses for the output of its calculated patterns;

(3) a program that does the reverse of (2);

(4) a defined metric for the similarity or difference between two patterns described in the language (1). That is a profoundly difficult problem in itself -- the comparison has to know what differences matter, and which don't. Also the metric has to reflect (measure) the relative desirability of the pattern it's applied to;

(5) a program to do the comparison and calculate metric (4) and decide whether it's close enough to the desired result;

(6) A program to modify the original antenna description, perhaps stochastically, into a new description to be fed to NEC; invoke NEC to calculate the new pattern; and then invoke the program (5) to calculate a new difference metric (4);

(7) software to do the iteration of all that and decide when to stop. It ought to remember what changes it has already tried, too, if (6) is not stochastic, and maybe even if it is;

(8) no doubt, lots more I'm not smart enough to foresee.

Parts -- not nearly all -- of (5), (6), and (7) might be done by particle swarm optimization, which you mentioned in an earlier thread and which I had never heard of before, or by many other approaches abetted by a couple decades of innovation.

Certainly it's possible. Sounds like several lifetimes of work. Worth it? That's for you to decide. How useful? You decide that, too.
-- Carl
NN5I is offline   Reply With Quote

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Powered by vBadvanced CMPS v3.2.3

All times are GMT -5. The time now is 05:23 AM.

Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2018, vBulletin Solutions, Inc.