Biograph: Dr. Xiaoji Niu is a Professor of GNSS Research Center at Wuhan University in China. He got his Ph.D. and bachelor degrees from the Department of Precision Instruments at Tsinghua University in 2002 and 1997 respectively. He did post-doctoral research at the University of Calgary and worked as a senior scientist in SiRF Technology Inc. Dr. Niu leads a multi-sensor navigation group focuses on GNSS/INS integrations, low-cost navigation sensor fusion, and its new applications. Dr. Niu has published more than 90 academic papers and own 28 patents.

Title:A Proposed Universal Architecture for Multi-source Deeply-coupled GNSS Receiver

Abstract: With the advances of GNSS receiver technique and the trend of multi-source integrated navigation,

GNSS/INS deep integration, i.e. INS-aided GNSS baseband, and/or visual-aided GNSS acquisition/tracking have been developed and are considered for production. Meanwhile, with the deployments of multiple GNSS constellations, vector tracking GNSS receiver has been reconsidered for the benefit of joint tracking of all satellite channels.

This paper proposes an architecture for multi-source deeply-coupled GNSS receiver. The design starts from the conventional scalar-based baseband and a cascaded vector-based baseband architectures. Each satellite channel is tracking by Kalman filter instead of PLL, to get the benefit of inherent adaptability for signal degradations. Once the INS-aided carrier Doppler feedback is introduced, the scalar-based and the vector-based architectures converge to each others and become similar. The only difference left is to keep (for scalar) or remove (for vector) the carrier frequency feedback control to the NCO within each local channel. Although the scalar-based deeply coupled baseband keep the local channel Doppler feedback of the NCO, such feedback will be mainly taken over by the INS Doppler aiding once the INS has been initialized and converged. Simulation tests using GNSS hardware simulator have been made to compare these two architectures and verify the feasibilities. Results show that both of them can handle the signal attenuation/blockage of partial satellite channels by keeping a kind of virtual tracking status of the weak channels, and can improve the tracking sensitivity in general when all the satellite signals are attenuated. Considering that the scalar-based baseband has the capability of signal acquisition, it is proposed as a universal architecture of GNSS/INS deeply-coupled integration. The INS-aiding channel in the proposed architecture plays an important role that absorb, condense, and carry the navigation information from every satellite channels, and then distribute it back to each satellite channel. Therefore it is highly recommended to use this INS-aiding channel to carry other sources of navigation information, such as camera visual navigation, LiDAR SLAM, signal of opportunity positioning, etc. Instead of using these multiple information sources to aid GNSS baseband directly, which suffer the spatial and temporal mismatch issues, it is much more decent to use them to update the INS first, then to aid the GNSS baseband indirectly through the INS-aiding channel. The proposed a universal architecture of GNSS/INS deeply-coupled integration that can take the advantages of both scalar-tracking (e.g. acquisition capability) and vector-tracking (e.g. sharing information between satellite channels) by using the INS-aiding mechanism in a decent way. This architecture is also adaptive for aiding other navigation information into the system, which is appropriate for the coming multi-source or all-source position and navigation requirements.