[ntp:questions] GPS Weakness Could Sink Wireless

Terje Mathisen terje.mathisen at tmsw.no
Fri Dec 13 07:58:03 UTC 2013


Jan Ceuleers wrote:
> Interesting Light Reading article on the degree to which infrastructure
> (in casu wireless networks) is dependent on GPS timing signals, how
> little is needed to jam GPS (intentionally or otherwise), and what the
> impact of such jamming would be.
>
> It also talks about how PTP might or might not mitigate some of these
> issues.
>
> http://www.lightreading.com/mobile/mobile-security/were-jamming-gps-weakness-could-sink-wireless/d/d-id/706895
>
I spent 2.5 years as the architect of a national cell phone network 
replacement project so I do know a little about this issue:

The core do use GPS to sync up the central time servers, but they also 
have Rb local osc, which means that even if you were able to jam the GPS 
reception in all the areas that have reference gps receivers, you would 
also have to wait long enough so that the Rb clocks would drift apart.

For the GSM/3G/4G networks we have we don't need to know exactly what 
the time is (unlike the US CDMA setup which do need single-digit us time 
sync), we only need to keep the frequencies the same, and these only 
need to match up at the 20 ppm level in order to be able to do seamless 
base station handovers at bullet train speeds.

20 ppm is a _lot_ compared to the stability of an Rb osc, it would take 
at least some months before things got that bad, and if you had an 
adversary who could do wide-area GPS denial over several months, without 
getting detected and caught, then you have a _real_ problem.

Terje
-- 
- <Terje.Mathisen at tmsw.no>
"almost all programming can be viewed as an exercise in caching"



More information about the questions mailing list