Research on Particle Filter Based on Neural Network for Receiver Autonomous Integrity Monitoring
Abstract: According to the
measurement noise feature of GPS receiver and the degeneracy phenomenon of
particle filter (PF), in order to alleviate the sample impoverishment problem
for PF, GPS receiver autonomous integrity monitoring (RAIM) algorithm based on
PF algorithm combining neural network was proposed, which was used to improve
the importance state adjustment of particle filter algorithm. The PF algorithm
based on neural network is analized. And the test statistic of satellite fault
detection is set up. The satellite fault detection is undertaken by checking
the cumulative log-likelihood ratio (LLR) of system state of GPS receiver.The
proposed algorithm was Validated by the measured real raw data from GPS receiver,
which are deliberately contaminated with the bias fault and ramp fault, the
simulation results demonstrate that the proposed algorithm can accurately
estimate the state of GPS receiver in the case of non-Gaussian measurement
noise, effectively detect and isolate fault satellite by establishing
log-likelihood ratio statistic for consistency test and improve the accuracy of
detection performance.
Keywords: Global Positioning
System (GPS); Receiver Autonomous Integrity Monitoring (RAIM); Particle Filter
(PF); Fault Detection; Neural Network
Author: Ershen Wang
Journal Code: jptkomputergg160181
