Printer FriendlyEmail Article Link

Positioning Application: What happened on January 25-26 2016?

Environment/Versions
  • GPS
Answer
On 25 - 26 January 2016, GPS users encountered a rare event. Multiple satellites started to broadcast incorrect timing information, specifically the offset between GPS and UTC (USNO). The incorrect data was spread across three words in page 18 of subframe 4 of the GPS L1 C/A code (see Figure 1).    
           
                                                          Figure 1: Subframe 4, Page 18 of GPS L1 C/A code
 
The parameters of words 6, 7 and 8 allows a receiver to output correct UTC time. The information contained in these three words gives the offset between GPSST (GPS System Time) and UTC. Because GPS receivers rely on GPSST internally, this error would not have affected the positioning and navigation capabilities of receivers. However, timing receivers outputting UTC would have been affected.

Starting at 23:26:18 (UTC) on the 25th of January 2016 SVN 23 started broadcasting the incorrect data. A few hours later, 15 satellites were broadcasting the same incorrect data. The problem was an erroneous upload to the satellites. It wasn’t until 13:11:18 the next day that problem had been fully corrected. During this time several users reported seeing errors of up to 13.7 µs.

However, there was an indication, in the faulty data, that the receivers, reporting the error, ignored. The reference time for the faulty data was set to the 11th of May 2014, almost a year and a half difference from the actual date. This information is an indication of something being wrong with the data. At the same time, several satellites were broadcasting correct data, which should have been used instead of the faulty ones.

Spirent now offers a scenario where users can test the effects of this event, and make sure that their receivers are capable of handling such an event.

Product : PosApp