Estimation of GLONASS Carrier Phase Inter-frequency Biases

Because GLONASS employs frequency division multiple access technique, the inter-frequency biases (IFBs) generate in the double difference observations from different types of receivers. Therefore, it is significant to eliminate the IFBs for GLONASS high-precision positioning. The size of the phase IFBs rate is generally millimeter or centimeter, the IFBs of L1 and L2 are almost equivalent, and possess a certain linear relationship with the frequency number, which is conducive to the separation of the IFBs and the ambiguities. In this paper, two methods of phase IFBs correction are adopted to compare, including pre-calibration and real-time estimation. In the pre-calibration, the function model of the widelane observation residuals and the IFBs rate is established, and the IFBs rate is estimated by sequential estimation; in the real-time estimation, the GLONASS mathematical model considering IFBs is constructed, and the IFBs rate is estimated by Kalman filter. In the data preprocessing, the triple difference observations are used to construct slip test statistics for cycle slip detection. The LAMBDA method is used to fix the ambiguity. The experimental results show that the two methods can effectively estimate GLONASS IFBs rate and both the fixed rate and positioning accuracy are improved evidently after the IFBs elimination. The filtering method is more stable and suitable for real-time solution.