The invention relates to a resident trip OD matrix calibration method using mobile phone signaling big data and dynamic traffic assignment. The method comprises main steps: an urban road simulation network model is built, microwaves and bayonet data are used for calibrating simulation parameters of each road section in the built urban road network, a microwave detector is used for acquiring the real traffic flow in the actual road network, the mobile phone signaling data are used for acquiring the initial OD matrix traffic flow, the initial traffic flow is dynamically adjusted through a dynamic traffic assignment method to enable the initial traffic flow to be close to the real traffic flow in the road network, errors between the simulation and estimation traffic flow and the real traffic flow are calculated, the calibration is ended if a threshold is met, or otherwise, simulation-based dynamic assignment is continued until the calibration is ended. The mobile phone signaling big data are adopted to calibrate the resident trip dynamic OD matrix, and the method has the advantages of stable and objective data information source, large sample amount, wide coverage range, high data precision, strong dynamics and the like.