Victim Localization and Assessment System for Emergency Responders

Hyungchul Yoon, Reza Shiftehfar, Soojin Cho, Billie F. Spencer, Mark E. Nelson, Gul Agha

Research output: Contribution to journalArticlepeer-review

24 Scopus citations


In minor to moderate natural and man-made disasters, such as earthquakes and fires, people may be trapped inside buildings and hurt by the disaster. Considering that trapped victims may be unconscious, there is a high demand by emergency responders to get information on the locations and physical statuses of trapped victims inside a building during a disaster. In this paper, a smartphone-based, in-building emergency response assistance system, named iRescue, is presented. The system is comprised of two subsystems: a Victim Positioning System (VPS) and a Victim Assessment System (VAS). The VPS uses the received signal strength indicator of Wi-Fi signals from multiple wireless access points with referencing a pre-established Wi-Fi fingerprinting map of a building. The VAS uses patterns obtained from measured 3D acceleration changes by status of a victim. A Naïve Bayes classifier is employed for both VPS and VAS: for localization in between the fingerprinting map and for recognition of activities to be used for status assessment. The performance of the VPS has been validated by a localization test on a complex building. The VAS has been validated by activity simulation test with five people and real-time monitoring of a person equipped with an activity recording device.

Original languageEnglish
Article number04015011
JournalJournal of Computing in Civil Engineering
Issue number2
StatePublished - 1 Mar 2016


  • Activity recognition
  • Disaster rescue
  • Emergency response
  • Indoor localization
  • Naïve Bayes classifier
  • Smartphone


Dive into the research topics of 'Victim Localization and Assessment System for Emergency Responders'. Together they form a unique fingerprint.

Cite this