Conventional pedestrian dead-reckoning systems that use shoe-mounted inertial sensors are not suitable for use in daily life. For example, such systems are not convenient because they compel individuals to wear such pairs of shoes; moreover, these systems are not suitable for use in barefoot conditions. A novel gait logging system is proposed to estimate foot steps and walking trajectories using a combination of high frequency vibration and inertial measurements from the upper part of the ankle. First, a three-step method is proposed to estimate Heel Strike and Toe Off. It is verified that the present method is suitable for estimating Midstance (MSt) time. Then, a trajectory and position estimation procedure is proposed that considers the MSt ankle velocity. A model for MSt ankle velocity is derived. The performance of the trajectory estimation is evaluated with various types of footwear. It is verified that the proposed method is suitable for estimating step lengths of each stride and total distance regardless of the footwear. Furthermore, the trajectories estimated using the proposed method approximated the actual walking path with high accuracy.