Despite extensive investigation robust mobile indoor navigation remains a challenging problem. Many existing approaches require infrastructure (e.g., indoor beacons) to achieve reliable accuracy. In this project we exploit inertial sensors and only require indoor map information. Essentially our solution relies on accelerometer based step detection and step length estimation, along with magnetometer measurements for direction to estimate location within known indoor map constraints.