Telecommunications Science ›› 2019, Vol. 35 ›› Issue (6): 70-77.doi: 10.11959/j.issn.1000-0801.2019150

• research and development • Previous Articles     Next Articles

Real-time detection method of smoke based on Wi-Fi

WU Zhefu1,TANG Wanli1,CHEN Cheng2,FANG Luping1()   

  1. 1 College of Information Engineering,Zhejiang University of Technology,Hangzhou 310023,China
    2 China Comservice Huaxin Consulting Co.,Ltd.,Hangzhou 310052,China
  • Revised:2019-05-20 Online:2019-06-20 Published:2019-06-20
  • Supported by:
    The Natural Science Foundation of Zhejiang Provincial of China(LY18F010025);The Natural Science Foundation of Zhejiang Provincial of China(LY14F050004);The Natural Science Foundation of Zhejiang Provincial of China(LY13F010011)


Fire-related smoke real-time detection method is one of the important means to ensure public safety.A fire smoke detection method based on channel state information of Wi-Fi network was proposed to avoid the high cost of existing fire sensor system or the blind spot of video detection.The method need not require a dedicated fire smoke monitoring device.By studying the influence of fire smoke on wireless transmission signals,the smoke characteristic parameters in the channel state information were extracted and classified by machine learning,thereby realizing the real-time monitoring of environmental fire.The experimental results show that the accuracy of environmental smoke detection based on Wi-Fi channel status information can reach 98.06%.

Key words: wireless channel, channel state information, smoke detection, machine learning

CLC Number: 

No Suggested Reading articles found!