Assisted GNSS: An Open Source SDR-Based Approach

The Global Positioning System (GPS) was designed to require approximately 30-45s for its time-to-the-first-fix (TTFF). However, with the growth of its civilian applications, narrowing TTFF for GPS receivers is becoming a major topic of interest within the Global Navigation Satellite System (GNSS) community. This led to the creation of Assisted GNSS (A-GNSS), which allows receivers to narrow TTFF through the use of a smarter acquisition strategy [1] combined with the ephemeris data. Ephemeris data includes all data necessary for satellite positioning, and is transmitted by a cellular network in an effort to avoid delays. Without it, users are forced to wait for a full frame broadcasted from navigation satellites, which usually takes 30 seconds for the GPS L1 CA signal. Instead, A-GNSS can utilize the existence of previously established ephemeris data to provide instantaneous positioning. With current cellular data rates, it is possible for A-GNSS receivers to access and utilize ephemeris data with no delays. Since research in A-GNSS is becoming increasingly important, we propose a software-defined-receiver (SDR) toolkit based on A-GPS technique in this paper. The corresponding code can be downloaded at: https://www.colorado.edu/lab/gnss. Both the server and A-GPS receiver are illustrated and analyzed in detail. Specifically, in the server side, the proposed A-GPS SDR uses a full GPS SDR [2] to process a live GPS data and save the necessary ephemeris data. At the receiver side, we use the saved ephemeris data as the assisted data (aData) input, and perform the A-GPS receiver algorithm [1]. The receiver side SDR is based on the open source Matlab SDR proposed in [2]. The A-GPS SDR toolkit is accessible for both education and research purposes, by students and researchers interested in A-GNSS techniques. Additionally, we provide the multi-constellation interoperability of GPS and Galileo using A-GNSS technique. The two systems are combined in the pseudo range measurement level, and the GPS Galileo time offset is solved with the introduction of the additional state in the least square process. Experimental results show that a joint GPS Galileo A-GNSS receiver has a better positioning performance than a standalone A-GPS receiver because a better GDOP has been achieved when adding Galileo constellation.