PPRuNe Forums - View Single Post - gps spoofing - researchers feed false gps signals into a ships navigation system
Old 10th Aug 2013, 19:25
  #22 (permalink)  
underfire
 
Join Date: Aug 2013
Location: PA
Age: 59
Posts: 30
Likes: 0
Received 0 Likes on 0 Posts
I dont believe that is correct either.

ADS-B data comes from the IRU. Remember that GPS tells the aircraft where it was, not where it is. The latency in the GPS system is run through a Kalman filter and coupled with other data from the aircraft, such as airspeed.

The GPS signal itself, it 300 bits transmitted over 6 seconds. So, each satellite broadcast takes 6 seconds for the entire string. (think of how far the aircraft has traveled in 6 seconds)

The filters have the ability to weed out erroneous GPS signals or ones that dont match between broadcast as the aircraft is moving, based on the estimation of where it is.

The aircraft navigation systems need a certain number of sats depending on what you are doing. The minimum of 4 sats will give you horiz and vert location with no error checking. Most aircraft systems will require at least 5 sats for navigation with redundancy. If this minimum is temporarily not met, the IRU has some ability to span a gap, but them will go into a degraded mode with drift.

Surface ships are slow moving and don't require vertical, so fewer sats are necessary. The IRU is set up with a much larger outage time and IRU drift rates were never set up for precise navigation. Therefore, for a surface ship, it is far easier to broadcast 2 sat signals with a pseudorange correction because the ship is moving so slow, and doesnt likely have a very tight filter for the error corrections.

Trying to spoof an aircraft, you would have to follow it in the air, and replace the precise signal and pseudo range error that each individual sat has. If you are not able to do that, the aircraft will disregard the signal.

Back to aircraft and ADS-B. With the Kalman filter balancing the aircraft location, rapid changes in direction and speed will affect the position accuracy. Remember back when each sat signal takes 6 seconds, and you need at least 4, so, the latency of the estimation can be upwards of 30 seconds to get all of the sat signals, balance them and estimate with aircraft speed where the aircraft is. A rapid decrease in speed or a tight radius turn while descending on final can really throw things for a loop.

BTW, some ADS-B systems have their own independent GPS antennas and are not even coupled with the aircraft FMS
underfire is offline