Abstract: Different segments of the trajectory carry different directional information and moving patterns, which have symbolic significance in the behavior interpretation of moving objects.This paper proposes an approach of direction-quadrant mapping oriented trajectory moving pattern analysis.Moving patterns of the single trajectory are regarded as sequences of direction characteristics and state changes on the time axis. Firstly, the direction characteristic of trajectory segments is calculated by sliding window. Then the single trajectory is encoded as a sequence of characters based on the direction quadrant, on which the regular expression of the moving pattern is defined. Finally, the finite machine is used to match strings corresponding to specific moving patterns, so as to detect patterns of the single trajectory efficiently. Experimental results based on Wikiloc flight trajectories and self-collection trajectories show that the proposed method in this paper is robust to the single trajectory moving pattern analysis, and can efficiently identify typical 8-shaped spiral pattern.