A bit of History
At our local model flying club, one of the oldest and largest in New Zealand a neighbouring large commercial site has on occasion been jamming the public 2.4GHz band. This has caused significant problems for the club and resulted in several crashed models. So the question is, how do we detect that the jammer is operating and let everyone know. It is potentially a a safety issue and we don’t want to see the loss of expensive models or damage to surrounding property.
A Development Project
This prompted some experimentation and the development of a relatively simple 2.4GHz jammer detector for the flying field. As you would expect, we do not have access to a radio jammer and will not be building one. We do need to be reasonably sure that the detector will alert us to the jammer but not to nearby wifi and other RC signals, which have not caused a problem in the past. Past investigation has shown that the jammer signal is at an apparently lower signal level than typical RC signals, but covers the entire band and is very effective at jamming. Looking at signal levels and spectrum analysis is complex for a simple single purpose project.
So what I came up with uses a standard FrSky RC transmitter and receiver set. The transmitter will be located in a small box on the roof of the club-house and powered by a small solar panel. The transmitter sends a brief signal every 2 minutes which the receiver mounted with the weather station locks onto. Being about 500 meters away and only 3-5 meters above ground level, the weak signal provides a RC link that is marginal compared to a in-flight link.
The receiver has a somewhat directional antenna aimed at the club-house which fortunately in this situation sees the jamming source in the background. Should the receiver miss 2 consecutive transmissions a notification will be posted on the club web site indicating that radio conditions are not good enough to fly. The only known caused of this in the past has been the 2.4GHz jamming.
Being somewhat directional it focuses on the known jamming source. It doesn’t rely on signal level, telemetry RSSI or spectrum analysis, but rather on a simple yes/no RC connection. The RC link used is one of the most popular and thought to be one of the best currently available. It a frequency-hopping system that uses the entire 2.4GHz band.
The Hardware
This is built as a one-off at the moment but could easily be converted to a dedicated PCB design. It is built around an existing circuit board.
The FrSky transmitter is a standard DHT (diy) module connected to a board with a PIC micro-controller. The PIC powers-on the transmitter and provides a PPM stream of 8 channels. Average current draw is about 4mA. As configured, the transmitter is only on for 5 seconds every 2 minutes. The antenna is a simple stripped coax, the tuype attached to many receivers. It can be oriented to reduce or adjust the apparent received signal.
The receiver is a miniature 8-channel module attached to a PIC micro-controller. When the transmitter is off the receiver outputs failsafe positions on the 8 channels. The PIC looks at one of the channels for the movement from failsafe to a specific position when the transmitter activates. This happens quickly. The PIC must see the channel position stable for one second or it doesn’t count. A signal that is too weal does cause the channel to jitter and can indicate a failed connection. The PIC must miss two consecutive connections before it indicates a failed connection; likely that the jammer is operating.
The firmware for the transmitter and receiver is most cut and paste from existing code and put together using MPLAB-X. Both ends are running a simple add-on to my standard real-time base code structure which I have been using for 20+ years. It runs at 4MHz, ‘s written in PIC assembly and occupies a tiny amount of memory.
If the transmitter code was modified to enable sleep, the biggest long term current draw would be the voltage regulator; which could also be eliminated.
Uploading to a web site
I haven’t done this yet, but it will probably involve a Orange-Pi zero with the receiver to simply ftp a couple of images to the server; a YES image or a NO image. This is pretty much how the weather station works. The O-Pi-Z can connect to the same wifi access point that the weather station uses.