Predicting Floor-Level for 911 Calls with Recurrent Neural Networks and Smartphone Sensor Data

Anonymous

Nov 03, 2017 (modified: Nov 03, 2017) ICLR 2018 Conference Blind Submission readers: everyone Show Bibtex
  • Abstract: In cities with tall buildings, emergency responders need accurate floor-level location to find 911 callers quickly. We introduce a system to estimate a victim's floor-level via their mobile device's sensor data in a two-step process. First, we train a Recurrent Neural Network (RNN) using Long Short-Term Memory (LSTM) hidden units to determine when a smartphone enters or exits a building. Second, we use a barometer equipped smartphone to measure the change in barometric pressure from the entrance of the building to the victim's indoor location. Unlike impractical previous approaches, our system is the first that does not require the use of beacons, previous knowledge of the building infrastructure or knowledge of user behavior. We demonstrate real-world feasibility through 63 experiments across five different tall buildings throughout New York City. Our system predicted the correct floor-level within +-1 floor with 91% accuracy
  • TL;DR: We used an LSTM to detect when a smartphone walks into a building. Then we predict the device's floor level using data from sensors aboard the smartphone.
  • Keywords: Recurrent Neural Networks, RNN, LSTM, Mobile Device, Sensors

Loading