Is a Modified Dipole Antenna Feasible for a Quad-Band GSM Design?

Click For Summary

Discussion Overview

The discussion revolves around the feasibility of designing a modified dipole antenna for a quad-band GSM application in a vehicle. Participants explore various antenna designs, their characteristics, and the challenges associated with integrating them into a vehicle's structure, particularly focusing on low-profile and omnidirectional requirements.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested
  • Homework-related

Main Points Raised

  • One participant proposes using a modified dipole antenna with parasitic elements to resonate at GSM frequencies (850/900/1800/1900) and seeks insights on feasibility.
  • Another suggests a log periodic antenna but notes its broadband nature may introduce noise outside the desired frequencies, recommending a segmented dipole instead.
  • Some participants discuss the necessity for the antenna to be omnidirectional due to the nature of cell coverage and the need for it to cover a wide horizontal area.
  • There is a suggestion to use a vertical monopole antenna with the vehicle's metal body as a ground plane, while also considering the integration of the antenna into the windscreen for aesthetic purposes.
  • Participants mention the potential use of inductive elements in antenna design to achieve resonance at higher frequencies.
  • Questions arise regarding the use of EMPIRE software for simulations, with requests for tutorials and STL files for car bodies.
  • There is a discussion about S-parameters in antennas, with participants clarifying their relevance to measuring resonant frequency.
  • Concerns are raised about the legality of transmitting on certain frequencies and the potential for interference with emergency services.

Areas of Agreement / Disagreement

Participants express multiple competing views on the best antenna design for the project, with no consensus reached on a single approach. The discussion remains unresolved regarding the optimal solution for the quad-band GSM antenna.

Contextual Notes

Participants mention limitations in their knowledge of certain antenna types and simulation software, as well as the need for further research into multi-band vertical antennas and antenna design theory.

Who May Find This Useful

This discussion may be useful for undergraduate engineering students, particularly those interested in antenna design, vehicle communication systems, and simulation software for electromagnetic applications.

AhmedEzz
I searched the forum and didn't find any input on this subject.

Dear PF members, I need your help. I am at training for a university, and am assigned a project where I should design a quad-band (or tri) GSM antenna for a vehicle. I am going to second year EE undergrad. so I am only required to set the furnishing ground for the project and produce a report at the end.

I have searched google, wikipedia, IEEE, and read Balanis and ARRL amateus book (first chapter only)...

I need to ask a few questions which I hope you might consider answering using your superior knowledge on the subject.

1 - I have in mind a "modified" dipole antenna which have parasitic elements. Each of those elements would resonate at one of the frequencies of GSM system (850/900/1800/1900). Is this feasible? If yes, can you give me an insight on how this can be done?

2 - I will be using EMPIRE software for simulations, is anyone familiar with it?

That is it for nowa, I hope you might not find this post too daunting for you. (after all, we are engineers).
 
Engineering news on Phys.org
You could use a log periodic antenna, but it is broadband, so you will also pick up noise in the band that is not on the 850/900/1800/1900 frequencies:

http://en.wikipedia.org/wiki/Log-periodic_antenna

A better choice would be a segmented dipole, where the inner segment section resonates in the 1800/1900 band, and the full sections resonate in the 850/900 band. You use inductive elements to separate the inner dipole sections from the outer ones, so that only the inner sections resonate at the higher frequencies.

There's also an interesting "Dual Band Blade Antenna" discussed at wikipedia.org

http://en.wikipedia.org/wiki/Dual-band_blade_antenna#Dual_Band_Antennas

If you use a simple diople, though, it will not be very directional. You might add parasitic elements (that are also segmented) to get some directional gain. I'm not that familiar with GSM antennas -- how do you aim them when they're on a vehicle?
 
they are supposed to be omni-directional (isotropic)...which makes sense because cells don't exist in one direction, they are everywhere. Therefore, it has to be omni-directional but it also has to cover a lot of horizontal ground.

It should also have a low-profile as security risks are high. Anyone can put a huge antenna in his car but the cleverness is to make it invisible.
 
AhmedEzz said:
they are supposed to be omni-directional (isotropic)...which makes sense because cells don't exist in one direction, they are everywhere. Therefore, it has to be omni-directional but it also has to cover a lot of horizontal ground.

Ah, sorry, I thought GSM was a satellite-based phone thing. If it's terrestrial, for a vehicle, I'd just use a vertical monopole, with the car metal body as the ground plane. Segment the vertical element with the inductive element placed to give you the higher frequency with the first element. You can google dual band inductive trap antenna for more ideas.
 
Nope, I want to integrate it in the windscreen. It is a novel thing, didn't find so many publishing on it. I am thinking rectangular microstrip antennas but I almost know nothing about them.

Your idea of a vertical monopole would have worked but as I said, it must be almost invisible.EDIT: You must be put off I didn't mention this before but I didn't want to narrow down my options before anyone shows interest.
 
Last edited by a moderator:
Un-visible, eh? Well, the quarter-wave combination element for 900MHz is about 8.3cm long (maybe less with the effective center load), so you could hide it in one of those low wings that you see on the rear trunks of some quasi-performance sedans. Just put it inside the vertical support of the wing on one side or the other... This is for your EE report, right? So you can set some creative rules, like you are designing the GSM antenna for sport sedans with wings (non-metallic wings, of course)... :approve:
 
Well thanks a lot , I thought you will run out of ideas and not reply. Still, its weird that non of the other PF members replied...51 views. Where the engineers and physicists in here?

lol, I guess I'm just asking for too much. Thank you very much berkeman for your help.
 
Wow, so there is something that you guys really don't know...never thought that might happen.

One question though and I hope someone would help me out a bit here.

Does anyone know "Empire XCcel" simulation software? If yes, do they know any tutorials there? Do you know where can I get STL files to import to the simulation (Car bodies for example?)
 
OK, I reckon my last questions were very specific and honestly hard.


However, this is an easy question that I hope you might help me with.

What is the S-parameter in antenna?
 
  • #10
AhmedEzz said:
What is the S-parameter in antenna?

"Scattering" parameters are used to characterize the input and output impedances and transfer functions of a 2-port network:

http://en.wikipedia.org/wiki/S_parameters

The antenna can be considered such a network (you can measure the imput Z and the transfer function, for example).
 
  • #11
The S I'm talking about is measured vs. frequency so that you can know your resonant frequency. If this is the same S then please tell me... (there's so many parameters called S)...lol
 
  • #12
AhmedEzz said:
The S I'm talking about is measured vs. frequency so that you can know your resonant frequency. If this is the same S then please tell me... (there's so many parameters called S)...lol

Yes, the S parameters are measured versus frequency. You treat the antenna like a 1-port network. See "antenna measurements" in this Network Analyzer Application Note from Agilent:

http://cp.literature.agilent.com/litweb/pdf/5965-7917E.pdf

You can also use a VSWR meter between a signal generator and the antenna to find the frequency where you have minimum/no return power, which means the antenna is resonant, and the input Z is real, since all the power input is radiating away.

BTW, be careful with antenna experiments -- if you start transmitting in a band that you are not licensed in, you can cause harmful interference (and violate the FCC laws). You wouldn't want to be stepping on the local Police or Fire frequencies accidentally. That would not be good.
 
  • #13
Thanks a lot for your help Berkeman. We have a say in my country (Egypt) " He who serves people most, is their master".
 
  • #14
Glad to help. BTW, I did PM a friend to look at your thread questions above, but I think he may be on vacation. You may still get some feedback from this thread.
 
  • #15
I really hope so because this project can change my whole future. If I could succeed and impress my professor, he might ask me to do more projects for him, which very much improves my career prospects and boosts my potential.
 
  • #16
Last edited by a moderator:
  • #17
berkeman said:
If it's terrestrial, for a vehicle, I'd just use a vertical monopole, with the car metal body as the ground plane. Segment the vertical element with the inductive element placed to give you the higher frequency with the first element. You can google dual band inductive trap antenna for more ideas.
This is what I am thinking too, an isotropic multiband vertical, using the vehicle's metal chasis as a ground-plane. With the bands (and therefore dimensions) you are working with, you can design a high gain antenna for this application. (One method of increasing gain of a vertical antenna is to design a collinear array).

You probably want to research multi-band vertical antennas. If I were doing this project, I would start by looking my key words up in the school library. I would want to look at several books on antenna design & theory. Also find an amateur radio (ham) club in your area and ask the members about multiband vertical antennas. You will be surprised at the wealth of information they can provide. Many of them, already use multiband antennas and can speak with you, from experience. There may even be an active club at your school. Here is a site with a few ham operators in you country. Egyptian Amateur Radio Assembly.

Regarding invisibility, I've heard of apartment/condo dwellers hiding their HF vertical antennas inside of a plexiglass flagpole and hang a real flag on the pole :biggrin:. Can you think of an adaptation of this concept, for a vehicle?
 
Last edited:

Similar threads

Boost Your Antenna Design Skills: Essential Resources for Microwave Antennas
  • · Replies 12 ·
Replies
12
Views
5K
Graduate BRS: Random Comments on Some Recent PF Threads
  • · Replies 94 ·
4
Replies
94
Views
14K