The use of commercial Inertial Measurement Units (IMUs) within navigation filter applications has seen a steady increase with the large growth of the smart phone market. Unfortunately, the IMUs utilized within mobile phones tend to be highly inaccurate and susceptible to large amounts of noise and bias. Indoor locations in particular are well known to poses various sources of magnetic interference which can cause magnetometer readings to be inaccurate at best. To counteract this, IMU based navigation systems often take advantage of application-specific characteristics to reduce the effects of noise on the resulting filter. In this paper, an application-specific navigation filter is presented specifically for indoor environments that reduce the dependence on the magnetometer during times of detected interference. The results of this filter are combined with a Pedestrian Dead Reckoning (PDR) system to provide a robust indoor localization solution.