Analysis and Verification to the Effects of NH Code for Beidou Signal Acquisition and Tracking

K. Yan, H. Zhang, T. Zhang, L. Xu, X. Niu

Abstract:	Beidou B1 and GPS L1 signal have similar characteristics in general, e.g. the periods of their spreading code are both 1ms. But for the data modulation, the data bit rate of GPS is 50bps which means that 1 bit data lasts 20ms and includes 20 spreading code cycles. The data bit rate of D2 code modulated in Beidou signal is 500bps which means that 1 bit data only lasts 2ms and includes 2 spreading code cycles. The data bit rate of D1 is 50bps originally, but after modulated by Neumann-Hoffman code (NH code) the data bit rate increases to 1kbps. So compare to GPS signal, the data bit rate of Beidou signal increases significantly. Particularly, the D1 code modulated by NH code has 1kbps data bit rate which makes data bit transitions become possible in every milliseconds. The widely used frequency discriminator of the frequency lock loop (FLL) in the conventional GPS tracking loop is four-quadrant arctangent discriminator which is optimal at high and low SNR and has a wide frequency pull-in range. But this discriminator is sensitive to data bit transitions, so two adjacent integration samples should be within the same data bit interval. In most situations the low data bit rate GPS receiver can meet this condition, therefore the probability that the data bit transitions affect the discriminator is relatively small and the FLL can work correctly. But in Beidou system that has a high data bit rate, the situation changed. For the D2 code, the data bits last 2ms, so every two pair of adjacent integration values has one pair of adjacent integration values in the different data bit intervals because of the data bit transitions. For the D1 code, the data bits last 1ms so every pair of adjacent integration values are in the different data bit intervals. Therefore the Beidou receiver should choose a discriminator that is data bit transitions insensitivity. Since the conventional GPS receivers normally use data bit transitions sensitivity discriminator, they can’t be used Beidou receiver directly without any modifying. On the other hand, since the Beidou system has not been established for a long time, there are few documents analyzing the FLL of Beidou receiver. By analyzing the principle of the frequency discriminator, this paper derived the relationship between the frequency error output by the frequency discriminator and the adjacent integration values. It illustrated that the four-quadrant discriminator may cause a wrong frequency error because of the data bit transitions, so that the receiver can’t correctly tracking satellite signals. On the contrary, the two-quadrant discriminator is suitable for Beidou receiver because it is insensitivity to data bit transitions. However the pull-in range of the two-quadrant arctangent discriminator is only half of the four-quadrant arctangent discriminator. It means that the receiver must acquire the Doppler frequency estimation with an error that is half of the four-quadrant discriminator, before tracking the Doppler frequency using the FLL. When the satellite signal exists and is detected by the decider, the difference between the estimated Doppler frequency value and the true Doppler value may be out of the FLL pull-in range which leads to the failed pull-in. This paper illustrates it through the effects of the frequency error to the integration magnitudes and the possibility density function of the integration magnitudes under three circumstances, which are no signal (only noise), low signal-to-noise ratio (SNR) signal and high SNR signal. Therefore the precise Doppler frequency estimation needs to be done after the signal is detected, and then the signal loop can switch to the frequency tracking state. The analysis was verified by real tests using both software and hardware receivers. The intermediate frequency (IF) data of Beidou were sampled can saved as file, then processed by software receiver. The test result shows that the receiver using the four-quadrant discriminator can not track the satellite signal correctly; the tracking Doppler frequency has an error of 500Hz compare to the real Doppler frequency; the navigation data bits demodulated from the signal are wrong. Because the 500Hz frequency error in the four-quadrant discriminator is equal to 180 degree phase inverse when using 1ms integration time, the NH code series 11111_01100_10101_10001 is demodulated to the wrong series 10101_11001_11111_00100, but the tracking loop seems as if it tracks correctly and no losing lock phenomena can be detected. When switched to the two-quadrant discriminator combined with the precision Doppler frequency searching mechanism, the receiver can track the satellite signal and modulate the navigation message correctly. The results were further verified in the FPGA+DSP hardware receiver. The proposed method can acquire and track the Beidou signal stably and implement the positioning on the hardware platform. The tests in the software receiver and the hardware receiver show that the receivers for high data bit rate GNSS systems must adopts a data bit transitions insensitive frequency discriminator. Correspondingly, the tracking loop needs a more accurate initial frequency estimation value, so the acquisition mechanism should be modified accordingly. By analyzing the difference between GPS and Beidou signal structures and the corresponding different points for FLL designs, this paper provides a helpful reference to the Beidou receiver design.
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:	107 - 113
Cite this article:	Yan, K., Zhang, H., Zhang, T., Xu, L., Niu, X., "Analysis and Verification to the Effects of NH Code for Beidou Signal Acquisition and Tracking," Proceedings of the 26th International Technical Meeting of the Satellite Division of The Institute of Navigation (ION GNSS+ 2013), Nashville, TN, September 2013, pp. 107-113.
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In