Detection and Mitigation of Errors on an Ultra-Tight Integration System Based on Integrity Monitoring Method

F. Qin, X. Zhan, X-L. X. Zhang

Abstract:	In recent decades, GNSS/INS integrated navigation systems have been developed since the strengths and weaknesses of GNSS receivers and INS uniquely compliment each other. Generally, the three architectures of integrated navigation systems are loose integration, tight integration, and ultra-tight integration, listed in order of complexity. Nowadays, ultra-tight integration is increasingly becoming important due to its numerous advantages in terms of the robustness in high dynamics and the immunity to jamming and interference. In this method, the dynamics on the GNSS signal are measured and estimated through IMUs (Inertial Measurement Units), and then the information can be feedback to the receiver tracking loops to compensate the vehicle dynamics, resulting in significant reduction in tracking bandwidth to the noise suppression. At present, the interference and spoofing have a serious influence on the GNSS signal since it is vulnerable and easy to be interfered. Though the ultra-tight integration can improve the immunity to jamming and interference by integrating all tracking loops into a Kalman filter, the interference is still not fully suppressed, particularly in the environments where the visible satellites are less and the interference is serious. In this case, a method to detect and mitigate the interference is proposed based on integrity monitoring in this paper. By this method, the tracking loop interfered and spoofed can be detected and removed from the ultra-tight integration filter. After the interference is mitigated, the more accurate ultra-tight integration navigation can be achieved. The integrity monitoring of GNSS/INS integrated navigation systems has been largely investigated since it can be implemented with less visible satellite compared to traditional RAIM (Receiver Autonomous Integrity Monitoring). In these researches, the Kalman filter residual is used as test statistic to detect step interference while its rate is used for the detection of slowly growing interference. By these methods, the interference on the ultra-tight integration can be detected, but which channel interfered is uncertain. It is limited for the GNSS interference mitigation. Hence, to solve this problem, a novel integrity monitoring algorithm is proposed in this paper. In this monitoring, the pseudorange and pseudorange rate residuals from every tracking loop are used as test statistics in this paper to analyze and determine which tracking channel is interfered. Moreover, both the step and slowly growing interferences can be detected rapidly. The pseudorange and pseudorange rate residuals computed from the code and carrier discriminators in each tracking channel of the ultra-tight integration are normally distributed with zero mean if there is no interference. Once a certain tracking channel is interfered, their distributions in this channel will become non-central normal distributions with non-zero mean. Hence, the pseudorange and pseudorange rate residuals in each tracking channel are used as test statistics to detect the interference on this channel. According to their distribution characteristics, a binary hypothesis is build and the detection threshold is calculated from the probability of false detection and missed detection employed. The threshold is the evaluation criterion whether the interference exists. Through this statistic analyses in all channels, it can be determined which channel is interfered. The test statistic of the pseudorange residual is used to detect step interference, while the test statistic of the pseudorange rate residual is used to detect slowly growing interference. If the interference is detected and the interfered channel is determined, the pseudorange and pseudorange rate residuals of the channel are eliminated from the measurements of the Kalman filter in the ultra-tight integration system. This method is called the interference mitigation. Once the interference cannot be detected again, the mitigation is canceled and the pseudorange and pseudorange rate residuals of the channel are re-added into the measurements. The detection and mitigation of the interference are performed in every tracking of all channels. Through this interference detection and mitigation, the navigation parameters can be exactly estimated and the tracking loops have better performances. To verify the interference detection and mitigation of the ultra-tight integration based on integrity monitoring, a high dynamic scenario including velocity and attitude change is simulated and an experimental platform is set up. According to this scenario, the GNSS constellation is built and the satellite RF (Radio Frequency) signal containing step and slowly growing interferences is transmitted from a HwaCreatTM GNSS signal simulator. Then, the RF signal is collected by a RF frontend (MAX 2769) and converted to an IF (Intermediate Frequency) digital signal for the process of a software ultra-tight integration receiver including the interference detection and mitigation algorithm designed. By the receiver, the GNSS interference can be detected and mitigated in tracking and navigation solutions. Furthermore, the tracking and navigation performances of the ultra-tight integrations with and without the detection and mitigation of the interference are compared and the advantages of this algorithm are proved. Finally, it can be concluded that the ultra-tight integration with the detection and mitigation algorithm has perfect performances even in high dynamic and serious interference environments. It is very valuable to many tactical applications. This paper is organized as follows: Section 2 describes the principle and structure of the ultra-tight integration. Section 3 proposes the detection and mitigation of the interference on the ultra-tight integration system based on integrity monitoring. Section 4 determines the test statistics and calculates the detection threshold according to their distributions. In Section 5, a high dynamic scenario is build and some experiments are performed to prove the detection and mitigation algorithm.
Published in:	Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013) September 16 - 20, 2013 Nashville Convention Center, Nashville, Tennessee Nashville, TN
Pages:	2102 - 2113
Cite this article:	Qin, F., Zhan, X., Zhang, X-L. X., "Detection and Mitigation of Errors on an Ultra-Tight Integration System Based on Integrity Monitoring Method," Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013), Nashville, TN, September 2013, pp. 2102-2113.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In