A Framework for Bayesian Receiver Autonomous Integrity Monitoring in Urban Navigation

Abstract

: A Bayesian framework for RAIM designed to meet the requirements of today's urban navigation is introduced. The framework is general in the sense that when contaminated observations can be modeled, integrity information can be computed. We consider the case of positioning in the presence of multipath signals. The proposed method consists of a Bayesian filter that computes the posterior distribution of the history of the presence of multipath signals and the sizes of multipath biases, along with the navigation parameters. From the posterior distribution, optimal integrity information can be theoretically derived. For practical implementation, the optimal Bayesian filter is approximated with various techniques, and suboptimal integrity information is achieved. Simulations show that the approximations of the Bayesian RAIM framework have good performance compared to the traditional least squares RAIM, the drawback being the computational complexity of the methods.