The whole angle mode gyroscope (WAMG) is considered to be the next generation architecture, but it is suffered from the asymmetry errors to conduct real products. This paper proposes a novel high frequency injection based approach for the error parameters online identification for the WAMG. The significance is that it can separate physical and error fingerprints to enable online calibration. The nonlinear WAMG dynamics are discretized to meet the requirement of numerical precision and computation efficiency. The optimized estimation methods are then constructed and compared to track asymmetry error parameters continuously. In the validation part, its results firstly prove that the proposed scheme can accurately identify constant asymmetry parameters with an overall tracking error of less than 1 ppm and the extreme numerical convergence can reach 10−12ppm. Under the dynamic asymmetry variation condition, the root mean square errors (RMSE) indicate that the tracking accuracy can reach the level of 10−3, which shows the robustness of the proposed scheme. In summary, the proposed method can effectively estimate the WAMG asymmetry errors online with satisfied performance and practical values.