Abstract: | Nowadays GNSS receivers are omnipresent navigation sensors even for personal positioning. In many cases, especially in E911, GNSS receivers are not working in ordinary times, but needed to find a position as soon as pushing the button in emergency. Time to first fix (TTFF) is dominantly dependent on the signal acquisition time. If receivers are not used in ordinary times or if receivers start without almanac or initial position information, the receivers should take cold start and search all the candidate PRN numbers, Doppler shift and code phase shift. In this case, it needs too much signal acquisition time. So we proposed a new technique for fast signal acquisition in NTM2006. In this paper, we’ll introduce more developed results than our previous research, especially on signal-receiving time, 10 times faster. Currently there are two kinds of signal acquisition technologies. One is a sequential searching method used in common GNSS receivers. It spends 1ms for each cells, and the candidates are totally L*M*N cells, which are LDoppler bins, M-code phase shift and N-PRN numbers. Therefore, it needs long time (usually over 1 minute) to search all the candidate cells. The other is a DFT (Discrete Fourier Transform) method used in software GNSS receivers. It stores sampled IF signals for code period (usually 1ms), and takes Fourier transform of these signals. Then the correlation values are acquired using DFT and inverse DFT process. It reduces acquisition time dramatically (about 3-5 seconds) but needs high speed (about 5~7 MHz) storage apparatus, which can not be implemented in current receivers. Therefore, it can be used only for future software receivers, which require different hardware architecture from the current one. Our technique, which is presented in NTM2006, takes advantages of above two algorithms and can be implemented in the current receiver architecture with very fast acquisition time (below 5 seconds). It stores the correlation measurements (1-ms accumulated I and Q) for only one PRN code changing Doppler shift and code phase shift in L*M cells. This process can be implemented in the current (H/W) receivers because it requires only 1ms-correlation measurements I and Q instead of all sampled signals. Then using this correlation data, the method can acquire Doppler and code phase shiftof all the other candidate PRN signals. Therefore, this method doesn’t need high speed sampling and storage equipments like software receivers, but reduces signalreceiving time (1-2 sec) for acquisition as compared with a sequential searching method. Our following research concentrated on the signal-receiving time. 1-2 seconds’ signal receiving time means we need to wait that much to start acquisition process. But now, we reduced the time as one tenth as before. We can search PRN, code shift and doppler offset with only one strip of cross-correlation sequence. So we need just 0.1 seconds' signal-receiving time. We will show simulation and experimental test results. |
Published in: |
Proceedings of the 19th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2006) September 26 - 29, 2006 Fort Worth Convention Center Fort Worth, TX |
Pages: | 1288 - 1296 |
Cite this article: | So, H., Jun, H., Kee, C., "A New Three Dimensional Signal Search & Acquisiton Algorithm Based on Cross-correlation Sequence with 0.1 Seconds' Signal Receiving Time," Proceedings of the 19th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS 2006), Fort Worth, TX, September 2006, pp. 1288-1296. |
Full Paper: |
ION Members/Non-Members: 1 Download Credit
Sign In |