BeiDou Consumer Receiver Chips at Last

P.G. Mattos, F. Pisoni

Abstract:	Whilst initial interest in the Beidou system was limited by its regional nature, the strong upsurge of the Chinese market caused semiconductor companies to investigate further with a view to providing chips for that market. The growing interest was stymied however by the absence of a signal specification (ICD), so when work began in 2011 the only source of signal details was the reverse engineering done by Stanford(Gao) and DLR(Germany). This was, however, sufficient to allow the basic signal to be received from the one or two MEO satellites that had worldwide orbits. The GEO and IGSO satellites in the constellation are not visible from Western Europe. Signal tracking was possible on existing silicon, because that silicon had implemented memory codes for Galileo. While Beidou codes are LFSR Gold codes, they could be generated in software and stored in/replayed from the code memory, with minor difficulties caused by the codelength and rate being unique, but at least based on the same 1.023 multiple as GPS and Galileo. The carrier offset from GPS at 1561MHz caused some difficulties, initially handled for development by sending a dual channel RF-recording box (Primo-II from NSL) to China and collecting RF samples with many satellites. Development could then continue in Europe by replaying the digital samples through the baseband….but still with no specification(ICD), during 2012 only tracking was possible, no data decode or positioning. During 2012 the carrier offset was handled by adding an external RF front end chip and programming it for the Beidou frequency, and some progress was made towards data download by collaboration with a Chinese company who had access to the spec, so could write the decode software. The ST software tracked the signal and extracted the data-bits without any knowledge of their coding, and these raw data/bits/symbols were passed to the China-generated software for decoding and storage. Similarly, when the ST software wished to use a satellite position, it made a request to the China software, which returned the result. Finally, in the dying days of December 2012, the full Beidou ICD was released, and immediately the missing data-decode and ephemeris functions were written in ST Italy and integrated with the existing software, allowing a full Beidou demonstration on the existing Teseo-2 silicon. This allowed much faster development as effectively all the software was being written in one location, and could be tested on captured data or the few visible satellites as required. Further satellites were also launched, and in January occasionally 4 were visible from Europe. The majority of the actual positioning testing was done at a location in China, however, with the software developed in Italy, and new versions sent to China for testing by email at least daily, taking advantage of the 8 hours clock-offset in the working day. This was necessary to be able to test with a full constellation of satellites, only available in the Asia region. Already also in design was the next generation silicon, which would not require the external RF front-end, and could generate Beidou codes directly without using the Galileo memory-based code generator. Known as Teseo-3, its design had already reached the FPGA stage, with a software emulation also available, when the full ICD was published. This allowed a timely change from just signal level tests to full positioning and navigation tests. The next step after the design is ratified in these tests is to release it for manufacturing, which should occur in Q2/2013, with first silicon samples available in Q3.The availability of the Teseo-3 silicon is thus marginal for the September conference date, but its progress will be reported, and it should be seen in volume products in 2014. The full paper will centre on the Teseo-2 Beidou field tests, as this product is already in volume use for GPS/Glonass and Galileo ready, and will be used in volume for Beidou for about a year while Teseo-3 silicon is completed, tested, and migrated into user equipment, where it will represent a cost and space reduction being truly single-chip just as the GPS/Galileo and Glonass versions. For the first time these results will be for true positioning tests (static) and road tests (mobile), instead of the signal tracking results to which researchers have been limited in the years preceding the availability of the full ICD.
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:	398 - 402
Cite this article:	Mattos, P.G., Pisoni, F., "BeiDou Consumer Receiver Chips at Last," Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013), Nashville, TN, September 2013, pp. 398-402.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In