This chapter systematically studies the estimation methods in real-time kinematic positioning. It first establishes the mathematical foundation through least squares adjustment, with rigorous analysis of its statistical characteristics and geometric interpretation. The discussion then progresses to sequential adjustment techniques, where the approaches are formulated for time-independent scenarios, while a parameter estimation framework is developed for time-dependent cases. The core focus resides in Kalman filter theory, which not only elucidates the conventional Kalman filter derivation but also introduces an enhanced window-recursive estimation algorithm incorporating sliding window mechanisms. The chapter objectively assesses the dynamic adaptability, computational efficiency, and precision stability of these methods, ultimately establishing their complementary relationships in modern navigation system implementations.