GNSS Receiver Autonomous Integrity Monitoring Missed Detection Evaluation Method Based on Poisson Distribution

The probability of missed detection (PMD) is a crucial metric for characterizing the integrity performance of receiver autonomous integrity monitoring (RAIM). The time consumption of the traditional PMD evaluation method based on the exhaustive testing is impractical for an extremely low probability. However, the traditional time-saving PMD evaluation method based on the posteriori binomial distribution is insufficient in controlling Type I and Type II error simultaneously with small test samples. A RAIM missed detection evaluation method based on Poisson distribution is proposed to introduce the concept of dead zone. The acceptable test samples size, the allowable missed-detection samples, and the predefined both types of error probabilities are exploited to characterize the dead zone of verification, such that the required test samples can be greatly reduced. Simulation and real-world experiment have demonstrated that the proposed method can effectively control Type I and Type II errors, simultaneously, and reduce the required test sample size up to 22.98% when compared with binomial distribution-based method.