In a disaster situation such as an earthquake, building collapse causes severe damages to both people and property. To rescue the buried persons in the scene of accident, it requires a lot of manpower and time. Moreover, secondary damage can occur during the rescue operation by additional collapses. Currently technologies that utilize images, sounds and radio signals are mainly used to detect buried victims in disaster situation, However, they still require close access to the collapse site because the detection range of them is limited to small area. Because a rescuer has to get around a wide disaster scene while detecting victims, there is a difficulty in finding the buried victims by missing Golden Time. To solve this problem, in this paper we propose a stochastic method to estimate the location of buried persons in collapsing area using Wi-Fi signals. We assume that every victim in the disaster site has a smartphone. The method utilize only a few number of Mobile Wi-Fi access points (AP) to make it available to detect victims’ Wi-Fi signals in disaster situations. Our method can contribute to life protection technologies by providing the locations of buried persons for a quick rescue in case of a disaster.