Abstract: In this paper, we first analyze the influence of four types of clock jump categorized according to their impact on GNSS observables to time tag and carrier phase measurements. Then we make a comparison of two mainstream methods. The results indicate that, the precision of carrier-phase corrected method is mainly depended on the precision of carrier phase, while the precision of time-tags corrected method is mainly affected by clock jump value. Then, we make an in-depth discussion about time-tags corrected method for the four types clock jumps, and provide an unified formula to construct carrier-phase-derived Doppler measurements. The testing is conducted with static data from IGS stations and kinematic data collected by car campaign by using conventional method and the proposed method. The results show that, receiver velocity suffered significantly from clock jumps, and the impact on static velocity estimation reaches up to cm/s and dm/s for 30s sampling data and 1 s sampling data respectively, and the standard deviation reaches 25 cm/s for kinematic data.