Hypothetically (or maybe not), if you had all the tools like an oscilloscope, spectrum analyser, function generator etc and several engineering degrees although that knowledge is a bit rusty, what would be your approach to optimize a current piaware or radarcape setup?
There is little that can be done with the receiver itself. Modern electronics are getting smaller and more integrated. You will also have a hard time finding data sheets since they are behind NDA agreements.
There are some single digit improvements for things like gain settings, cooling, EMI reduction. This might be very interesting to some but is not where I would start since the improvements are so small.
The main improvement are gained from the antenna and the feed line. There are a lot of posts on the ADSB forum because people are seeing huge improvements (double digits) for something they can do without fancy electronic equipment. With some equipment you might be able to optimize a bit more.
It’s not my intention to design the new pro stick. However. I live in a very densely populated area with transmissions on every frequency one can think of, Including nearby GSM masts and this tower of terror with a height of approx 500 feet at 5 miles distance. I have the equipment and instead of trying if the next cantenna will improve the message/sec rate, I should be able to make precise measurements on cable losses, LNA effects, antenna position etc.
My area of interest is creating the best antenna, amp, location, filter etc configuration at the connector that feeds the pro stick. Gains and other parameters in the software or the stick can help, but nothing beats feeding the piaware with the best possible signal. Any directions on how to methods how to test are welcome as it has been some time I pushed the buttons on a scope.
First step is to survey the landscape, in terms of amplitude v frequency – you’re going to need a filter between the antenna and the SDR. The question is how narrow a filter will you need. You can do that survey with a wideband SDR and an external antenna, or with a spectrum analyzer and an external antenna. I’d use an external antenna that’s as close to what you expect to use as possible.
You can look elsewhere on these forums for for detailed frequency response plots for the FlightAware filter, and other filters such as SAW filters, and some cavity filters. The filter tradeoffs are over cost, insertion loss, and bandwidth. Lowest cost is the Flight Aware filter, also with low insertion loss. Bandwidth? Good, but SAW and cavity filters are sharper.
SAW (surface acoustic wave) filters are a little more expensive, are very sharp in terms of bandwidth, but tend to have higher insertion loss (around 2 - 3 dB).
A cavity filter is the most expensive. A well built and adjusted cavity filter will also have the best performance in terms of bandwidth, and low insertion loss, often under 1 dB.
Applying those frequency response curves to your environment will help you determine how far to go as far as filtering goes.
It will also help you determine how good your feedline needs to be (in terms of leakage for example), and if your SDR (and Pi) needs to be in a metal box or not.
Let us know what you find, and what you do to get an operating ADS-B system – ask questions and we’ll be happy to help as we can.
I hope you decide to experiment, collect data, and make decisions based on that data, and share with us.
Another approach, just as valid, is to decide with your wallet – how much money am I willing to spend on a filter? Get the best for your budget and see how it works.
here you find all what you need:
great receiver: shop.jetvision.de/epages/648079 … s/66006%2B
great omnidirectional antenna: shop.jetvision.de/epages/648079 … ts%2F67020
great directional antenna: amateur-radio-antenna.com/is … 3-prof.php
great cavity filter: shop.jetvision.de/epages/648079 … ucts/71010
great amplifier/filter (saw or ceramic): store.uputronics.com/index.php? … duct_id=50
the above should help to max your site
p.s. test your frequencies at your site kmkeen.com/rtl-power/