Research on Processing Method of Low-frequency Time-code 1PPS Based on Kalman Filter

Improving the timing accuracy of low-frequency time-code by digital means is the key to developing the low-frequency time-code timing receiver. Kalman filter can filter stationary random signals in the time domain. High-speed parallel operation is the characteristics of FPGA, in engineering, it can achieve the purpose of high precision, low cost, low power consumption, and small volume. In this paper, a processing module of 1 pulse per second (1PPS) of low-frequency time-code is designed based on the FPGA platform to realize the timing function. Meanwhile, by the Kalman filter algorithm to process the phase difference between the 1PPS of low-frequency time-code and the 1PPS of local oscillator. Through simulation and experiment, the result show that the Kalman filter can effectively improve timing stability and ensure timing accuracy, the standard deviation of the Filtering results can reach sub-millisecond, the difference between the mean of measured value and mean of filter values is only 3 µs and ten-thousand-seconds frequency stability of filter values is about 10-7. It can meet the needs of actual engineering and also provide the technical foundation for the digital low-frequency time-code timing receiver.