Identifying satellite interference due to GSM rebroadcast (2011) [pdf] | Hacker News Reader

Identifying satellite interference due to GSM rebroadcast (2011) [pdf]

data.satirg.org

50 pointsfanf214 comments

Tracing down RFI sources can be immensely challenging - I remember reading of one case where the tower itself became a radiating element and a source of RFI.

https://www.thebdr.net/articles/warstories/trenches/trenches...

Here is another where the source of the RFI was a totally unexpected source.

https://www.thebdr.net/articles/warstories/neighbors/bug.pdf

For an example of more malicious interference, see https://en.wikipedia.org/wiki/Captain_Midnight_broadcast_sig...

Even weirder and more wonderful, and still no one knows who it was: https://www.youtube.com/watch?v=tWdgAMYjYSs

"Bandwidth limit exceeded"

Here's the same PDF from the Internet Archive:

https://web.archive.org/web/20170930150048/http://data.satir...

I'm sorry for asking this, but can anyone tl;dr the slideshow? The satellite was blindly receiving and rebroadcasting signals from a misbehaving cell tower?

https://news.ycombinator.com/item?id=15371607

So the satellite was rebroadcasting the signal at a frequency where it could interfere with GSM? Does the satellite just rebroadcast whatever it receives, verbatim?

Block up-converter (BUC) at uplink site (in this case, likely part of a VSAT terminal or Satellite Internet "antenna") takes L-Band (950-1525mhz) input via cable, converts it to C-band uplink (5850-6425mhz) frequency, and transmits to satellite. Satellite then rebroadcasts verbatim at C-band downlink frequency (3400-4200mhz). Except that one of the inputs didn't have a cable or filter attached so it picked up a GSM signal at 936-960mhz. This caused interference on the satellite.

For further context, the reason for this slightly convoluted setup is that normal coaxial cables are really lossy at the high frequencies used to communicate with satellites, so most systems use lower frequency bands on the cable runs and convert the signals by a fixed frequency offset at the actual antenna. It just so happens that the block of frequencies used on the cable runs is similar to common mobile phone frequencies.

Now that was some debugging! Especially liked the part where they grabbed an old GSM to decode the signal. KISS at its best

Context? Why was the signal being rebroadcast? What was the point?

TLDR: Unterminated input port at satellite uplink site acted as antenna, picking up nearby cell tower by accident. Fix: put a terminator on the port.

(2011)

Added. Thanks!

Reply on news.ycombinator.com