中圖分類號： TN98 文獻標識碼： A DOI：10.16157/j.issn.0258-7998.200149 中文引用格式： 何維,，胡安琪,，田增山，等. 基于家庭WiFi的室內(nèi)漏水檢測[J].電子技術(shù)應(yīng)用,，2020,，46(9)：69-73. 英文引用格式： He Wei,，Hu Anqi，Tian Zengshan,，et al. Indoor water leakage detection based on home WiFi[J]. Application of Electronic Technique,，2020，46(9)：69-73.
Indoor water leakage detection based on home WiFi
He Wei,，Hu Anqi,，Tian Zengshan，Li Ze
Chongqing Key Laboratory of Mobile Communications Technology,，Chongqing University of Posts and Telecommunications,， Chongqing 400065，China
Abstract： With the improvement of wireless sensing systems and wireless network technologies, human-computer interaction, behavior recognition, and non-contact sensing detection based wireless signals has received increasing attention in wireless sensing. Aiming at the condition that water leakage accidents in home environment, this paper proposes a system for detecting water leakage in real time based commercial WiFi signals. It mainly uses the channel state information in commercial WiFi signals, captures signal changes caused by different characteristics of objects by receiving signals, and completes the three-step processing: data preprocessing, feature extraction, and machine learning classification to achieve the goal of real-time detection of leaks,，and innovate the traditional leak detection system. In order to prove that the proposed system′s classification model is suitable for any ordinary home environment, this paper executes the test in real environment at the end, and the outputted results verify the feasibility of the system.
Key words : CSI,；liquid detection；WiFi,；SVM
0 引言
近年來,，由于低成本、低功耗,、小尺寸的傳感設(shè)備的出現(xiàn),，無線傳感系統(tǒng)和無線傳感網(wǎng)絡(luò)(Wireless Sensor Network，WSN)迅速發(fā)展,，通過接收器通信,，監(jiān)視物理環(huán)境條件[1-2]。無線感知是無線網(wǎng)絡(luò)中一個新興的尖端研究熱點,。通過使用無線信號(如WiFi,、雷達、聲波,、射頻識別(Radio Frequency Identification,，RFID)等)對人與環(huán)境進行非接觸式感知，在健康監(jiān)護,、新型人機交互,、行為識別等領(lǐng)域有著廣泛應(yīng)用，輔助系統(tǒng)智能化,、人性化,。
隨著無線感知技術(shù)的出現(xiàn)，基于信道狀態(tài)信息(Channel State Information,，CSI)的無設(shè)備傳感系統(tǒng)也受到重視,，主要包括活動識別、跌倒檢測和生命體征監(jiān)測。WiFall[3]和RT-Fall[4]分別利用CSI幅度和相位差檢測跌倒動作某些特征的衰減,。PhaseBeat[5]和TensorBeat[6]使用CSI來估計單個或多個人的呼吸率,。在最近的無線感知研究Wi-Fire[7]和Wi-Metal[8]中，分別使用CSI數(shù)據(jù)來檢測火災(zāi)事件和金屬物體及腫瘤,。
