Jump to content
Sign in to follow this  
Bod

the old antenna pointing post

Recommended Posts

I have had some thoughts about the antenna pointing post back then and many years have gone since.

I have a new idea about using two or three recievers with divergent antennas having strong pointing charecteristics, and using their different RSSI voltages fed into differential amplifiers and then use that for pointing the antenna array.

This is analog tech and it would be extremely fast.

One of the receivers could be the video-receiving one, or a fourth one wich would be centered, could be used, possibly with a very high gain antenna.

Wouldn't that beat a micro-processor controlled system with one poor sweeping antenna wich would induce video-drop outs at the end points ?

Would it not track a fast target, even at close range with no processing overhead at all ? (no noticably delay or lag that is)

Terry is especially invited in this discussion but anyone else is too. (of cource)

Cost of those receivers is nothing, compared to what you can achieve :)

I will make a test in this spring, also I will upgrade my recording capability with an archos AV 500 :) at least hopefully :)

Bo

Edited by Bod

Share this post


Link to post
Share on other sites

Hi Bo,

I had the same idea as you have guessed long ago when I was making an auto tracking aerial. I have made it work with 2 patch aerials but it can get very jerky with the aerials wanting to jump off track at full speed. I think it could be filtered out but due to the then high price of receivers and aerials I went down the single receiver and aerial route.

There are no drop outs at the end points due to the aerial beam being

much higher than the sweep.

If you want to use a higher gain for the center aerial then you must be sure your tracking is good enough to keep it on target all the time or you will loose your advantage.

I will be interested in your results as I am going to return to this project as soon as I have finished my 4 way diversity receiver.

The main problems will be filtering out odd/incorrect RSSI signals due to noise or multi path and getting a noise free 360deg connection to the rotating parts.

Terry

Share this post


Link to post
Share on other sites

Hello Terry.

I have thought about that jerkiness you mention. If the vehicle is for instance flying by some parked cars, there may be some multipath issues. It may be when your receiver is near reflecting things as well. Even moist ground can wreak havoc sometimes, not to mention wires buried in the ground.

My club is at a former military EMP trial site and we sometimes got some problems with steering our models with our normal 35 Mhz transmitters, it is at 58.390591°N 15.588953°E. Come and have some Swedish Sushi wich the Germans tends to like, such as "Grav-Lax" and "Inlagd Sill", if you go up north)

To American readers I may as well confess we eat horses too, but only those who had to be killed anyway,because they had to go, so as we do here. (We do not stock-pile dead horses to be let rottening or otherwise being misused as food)

Our transition from 27 MHzAM in the past, to 35MHzFM nowdays has improved our hobby but not made it safe at all.

Most traces of military land usage are erased. I guess however, that kind of problems could be minimized by dampening and filtering the RSSI-voltages by "RC-couplings", wich I know you guys knows a bucket full of, wich I do not, since I'm not an Elecronic guy really. (Ask me how to realise an air-speed sensor or stall indicator instead)

If there are problems with multi-path interference, I guess it would take some good thinking to counter, no matter if you are trying to sweep with one antenna, or use this "differential" approach. One obvius sollution, would be moving the receiver a bit, at least as a try (lol).

I don't think you mentioned if the jerkiness caused much video drop-outs or if it was just a annoying thing to watch happening?

I think the most perfect design WOULD use one reciever with one antenna and advanced software running from a nice MCU such as a pic 16c84 or better, the Propeller (or something ). Smart software could predict the vehicle and "know" it could not possibly warp 180° in field of view in one second (unless you are thick enough to fly too close), so it would "track" the vehicle blindly for some time until it got a good signal again (by parabolic or perhaps linear extrapolation or some hokus pokus like that (j/k)). If it could not find a signal then, it would panic and sweep from endpoint to endpoint (preferably intelligent), to get a good signal to track normally.

However, really advanced software is hard to write generally (at least for most people including myself), and for a MCU particurlarly. (Using a PC is not an option any longer, at least not for me).

That is the reason I got back to analog thinking, at least for this problem.

I am also very interested to hear a report from your tracking arial project, could you post a forum link refenrence, or describe it directly ?

Or just tell me a start point to look it up. (i have looked a bit looked but I might have missed something vital).

All good to you,

Bo

PS

I got this propeller demo board today and I really encourage everyone of you guys to at least check out at Parallax.com, what it can do. (The demo board is a lot more expensive than the chip alone). Video-overlays seems to be easy to realise with this chip (there is user written code samples for it on their forum). This Propeller is arguable the Superman of the MCU's (Or I'm just plain stupid ?)

I like this gadget and I do not work at Parallax nor do I get any bucks from them (too bad).

DS

Edited by Bod

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×