**Abstract**

Bangladesh is in the front line of battlefield of disasters due to geographical location and global warming faced over 200 natural disasters in past 40 years and most of the disasters were cyclones. People need to be evacuated and rescued before a cyclone landfall. In current practice, multipurpose cyclone shelter (MPCS) provides short-term safety for the disaster victims in Bangladesh, where people are rescued after disasters which cannot ensure survival of lives. This study aims to develop a method for efficient evacuation and rescue to reduce death tolls in the events of disasters. This study used Wi-Fi scanner and smartphones to detect people. An inbuilt index that includes name, address, mobile number, photo, service set identifier (SSID), and media access control (MAC) of smartphone was developed for 90 registered participants. In this controlled experiment, few new participants turned on hotspot in every five minutes. A new index of people with MAC/SSID was developed in MPCS simulating an emergency. Missing people were detected by comparing inbuilt index and new index, and ordered them self-evacuation. This method captured 100% evacuees. Most importantly, the proposed method will reduce death tools because the people are rescued earlier to a disaster hits a specific area.

**Keywords:** Wi-Fi scanner, MAC/SSID, cyclone, disaster, hotspot, rescue, evacuation
