Izvorni znanstveni članak

https://doi.org/10.17559/TV-20240319001414

Research on the Method of Using Wireless Sensing to Detect Indoor People during Fire Based on Unmanned Aerial Vehicle

Lin Tang ; Key laboratory of Specialty Fiber Optics and Optical Access Networks, Joint International Research Laboratory of Specialty Fiber Optics and Advanced Communication, Shanghai University, Shanghai 200444, China
Chao Wang ; Key laboratory of Specialty Fiber Optics and Optical Access Networks, Joint International Research Laboratory of Specialty Fiber Optics and Advanced Communication, Shanghai University, Shanghai 200444, China *
Yinfan Ding ; Key laboratory of Specialty Fiber Optics and Optical Access Networks, Joint International Research Laboratory of Specialty Fiber Optics and Advanced Communication, Shanghai University, Shanghai 200444, China
Meng Zhou ; Key laboratory of Specialty Fiber Optics and Optical Access Networks, Joint International Research Laboratory of Specialty Fiber Optics and Advanced Communication, Shanghai University, Shanghai 200444, China

* Dopisni autor.

Sažetak

Though improving the efficiency of urban land use, high-rise buildings greatly increase the difficulty in search and rescue during fires. Therefore, effective search and rescue in high-rise building fires has always been a hot and difficult issue in emergency management. Starting from the characteristics of high-rise building fires and the difficulties in addressing them, this paper introduced unmanned aerial vehicle (UAV) technology into search and rescue operations in fires. Based on the influences of dynamic factors on wireless signal (WS) states in a static environment, it investigated the techniques for extracting and sensing wireless channel state features when indoor personnel waves for help and proposed a method for indoor personnel detection based on channel state information (CSI) sensing. In the article, a single source wireless device (common Wi-Fi router) is used as the detection terminal to collect the wireless channel state, and the signal sampling, support vector machine (SVM), and other technologies are combined to realize the wireless channel state sensing and classification. The outcomes of the experiment revealed that the suggested approach can effectively succeed in through-wall detection and sensing for indoor personnel calling for help in the event of a fire, with an accuracy of 94.6%. Incorporating the non-invasive, robust, universal, and low-cost characteristics, the method promises broad application values in improving the search and rescue efficiency of firefighters.

Ključne riječi

wireless signal, channel state information, state sensing, through-wall detection

Hrčak ID:

321915

URI

https://hrcak.srce.hr/321915

Datum izdavanja:

31.10.2024.

Posjeta: 9 *