Despite the omnipresent smartphones and navigation apps, indoor navigation continues to be a challenge for blind and visually impaired individuals (BVI). Global Positioning System (GPS), while being highly reliable outdoors, is not a viable solution for indoor localization. Wi-Fi and BLE have been the most common alternatives to provide localization. However, this has met with limited success owing to unreliable signals resulting from interference, blocks, multipath etc. Oftentimes, having a means to situate oneself in unfamiliar surroundings (albeit with some error) can be quite useful, especially for BVI individuals. This thesis paper explores the possibility of using only onboard MEMS sensors in smartphones to enable localization without any reliance on external signals. Such an “internal” approach promises to reduce infrastructure needs for external signals while reducing battery drain from communications. Specifically, this thesis describes how Pedestrian Dead Reckoning can be used for indoor positioning and proposes the use of accelerometer and gyroscope sensor values to calculate the distance travelled. It proposes using a Kalman Filter to accomplish sensor fusion of the inputs coming from accelerometer and gyroscope. Experimental evaluations conducted in an indoor setting with multiple individuals of varying heights shows the proposed approach can provide reasonable position estimates when external signals are absent.