Statistical model of local noise/multipath effects for integrity monitoring in the urban environment
Abstract
For many safety-critical applications, a serious problem of the existing global navigation satellite systems (GNSS) consists in their lack of integrity. Two different integrity monitoring approaches can be defined for ground transportation systems : “the passive integrity” and “the active integrity”. The passive integrity allows all measurements without rejection and simply lengthens the protection zone (i.e. the alert limit). This method assures that the local noise/multipath effects, GNSS signal attenuation and corruption (with bounded impacts on the vehicle position) will not cause unsafe positioning. The active integrity method or RAIM includes two functions : the detection of a position failure and the exclusion of the contaminated pseudorange(s) from the GNSS navigation solution. For both integrity approaches (passive and active), an important element of the GNSS statistical description is the model of local noise/multipath effects, which can be dominant in the urban environment. With other models (tropospheric, ionospheric, etc.), the local model provides us with a formalized definition of precision and integrity levels of 1D (train) and 2D (surface vehicle) positioning.