室內(nèi)穿墻場景下的無源人體目標(biāo)檢測算法
doi: 10.11999/JEIT190378
-
重慶郵電大學(xué)通信與信息工程學(xué)院 重慶 400065
Indoor Through-the-wall Passive Human Target Detection Algorithm
-
School of Communication and Information Engineering, Chongqing University of Posts and Telecommunications, Chongqing 400065, China
-
摘要:
穿墻場景下，由于墻體造成信號嚴(yán)重衰減，接收信號中目標(biāo)反射信號的能量大幅下降，接收信號淹沒在收發(fā)機(jī)直射信號和室內(nèi)家具反射信號中，難以檢測墻后目標(biāo)。針對上述問題，該文提出一種新穎的基于多維信號特征融合的穿墻多人體目標(biāo)檢測算法(TWMD)。先對接收到的信道狀態(tài)信息(CSI)進(jìn)行預(yù)處理以消除相位誤差和幅值噪聲，再利用CSI的時序相關(guān)性和子載波相關(guān)性從相關(guān)系數(shù)矩陣中提取多維信號特征，最后使用BP神經(jīng)網(wǎng)絡(luò)完成特征與檢測結(jié)果之間的映射。實(shí)驗(yàn)結(jié)果表明，該算法在玻璃墻、磚墻和混凝土墻環(huán)境的識別精度分別在0.98, 0.90, 0.85以上。根據(jù)所統(tǒng)計(jì)的4000個各類樣本的檢測結(jié)果，與現(xiàn)有基于單一信號特征的檢測算法相比，該文算法在對不同數(shù)量運(yùn)動目標(biāo)的檢測上，獲得了平均0.45的精度提升。
-
關(guān)鍵詞:
- 無源人體目標(biāo)檢測 /
- WiFi /
- 信道狀態(tài)信息 /
- 多維信號特征
Abstract:In through-the-wall scene, due to the serious attenuation of signal caused by wall, the energy of target reflection signal in the received signal decreases significantly and the received signal is submerged in the direct signal of the transceiver and the reflection signal of indoor furniture, making the target behind wall is hard to be detected. In view of the above problems, a novel Through-the-Wall Multiple human targets Detection (TWMD) algorithm based on multidimensional signal features fusion is proposed. Firstly, the received Channel State Information(CSI) is preprocessed to eliminate the phase error and amplitude noise, and the multidimensional signal features are fully extracted from the correlation coefficient matrix by using time correlation and subcarrier correlation of CSI. Finally, the mapping between features and detection results is established by BP neural network. The experimental results show that the recognition accuracy of this algorithm in the environment with glass wall, brick wall and concrete wall is above 0.98, 0.90, 0.85, respectively. According to the detection results of 4000 samples, compared with the existing detection algorithms based on single signal feature, the proposed algorithm achieves an average accuracy improvement of 0.45 in the detection of different number of moving targets.
-
表 1 本文所提基于多維特征的目標(biāo)檢測算法
輸入：天線1的CSI接收矩陣${{{H}}_1}$，天線2的CSI接收矩陣${{{H}}_2}$，天線3 的CSI接收矩陣${{{H}}_3}$。 輸出：輸出特征${{F}}''$。 初始化： 天線個數(shù)$N{\rm{ = }}3$;輸出特征${{F}}''{\rm{ = 0}}$。 算法步驟： (1) for $i$=1, 2, $···,N$ (2) 用式(4)校正${{{H}}_i}$的相位，得到${{\theta }}$； (3) 用離群值刪除與小波去噪對${{{H}}_i}$的幅值進(jìn)行預(yù)處理，得到
$\left\| {{\tilde{ H}}} \right\|$；(4) 用式(6)、式(7)計(jì)算$\left\| {{\tilde{ H}}} \right\|$的相關(guān)系數(shù)矩陣${{A}}$； (5) 對${{A}}$進(jìn)行矩陣分解，得到第1個和第2個大特征值
${\lambda _1},\,{\lambda _2}$；(6) 用式(8)計(jì)算${{\theta }}$的相關(guān)系數(shù)矩陣${{C}}$； (7) 對${{C}}$進(jìn)行矩陣分解，提取第1個和第2個大特征值
${\gamma _1},\;{\gamma _2}$；(8) 用式(10)計(jì)算子載波相關(guān)系數(shù)矩陣${{S}}$； (9) 對${{S}}$進(jìn)行分解，提取前3個大特征值對應(yīng)的特征向量
${{{e}}_1},\,{{{e}}_2}, \,{{{e}}_3}$；(10) for $j$=1, 2, 3 (11) 用式(11)計(jì)算特征向量1階差分均值${\phi _j}$； (12) end for (13) 對$\left\| {{\tilde{ H}}} \right\|$的分布標(biāo)準(zhǔn)化得到${{Z}}$； (14) for $k$=1, 2, 3 (15) 將${{Z}}$投影到${{{e}}_k}$上，得到${{{p}}_k}$； (16) 計(jì)算${{{p}}_k}$的方差，得到${\beta _i}$； (17) end for (18) 構(gòu)建樣本空間${{{F}}_i}$； (19) 將${{{F}}_i}$輸入到神經(jīng)網(wǎng)絡(luò)模型，得到輸出特征${{{F'}}_i}$； (20) end for (21) for $l$=1, 2, $···, N$ (22) ${{F''}} = {{F''}} + {{{F'}}_l}$； (23) end for 
下載: 導(dǎo)出CSV
表 2 收發(fā)機(jī)參數(shù)設(shè)置
參數(shù) 發(fā)射機(jī) 接收機(jī) 模式 Injection Monitor 信道編號 149(5.749 GHz) 帶寬 40 MHz 發(fā)包速率 500 包/s 子載波個數(shù) 30 子載波編號 –58, –54, ···, 54, 58 發(fā)射功率 15 dBm
下載: 導(dǎo)出CSV
-
ADIB F, MAO Hongzi, KABELAC Z, et al. Smart homes that monitor breathing and heart rate[C]. The 33rd Annual ACM Conference on Human Factors in Computing Systems, Seoul, Korea, 2015: 837–846. doi: 10.1145/2702123.2702200. ZHANG Dongheng, HU Yang, CHEN Yan, et al. BreathTrack: Tracking indoor human breath status via commodity WiFi[J]. IEEE Internet of Things Journal, 2019, 6(2): 3899–3911. doi: 10.1109/JIOT.2019.2893330 ABDELNASSER H, HARRAS K, and YOUSSEF M. A ubiquitous WiFi-based fine-grained gesture recognition system[J]. IEEE Transactions on Mobile Computing, 2019, 18(11): 2474–2487. doi: 10.1109/TMC.2018.2879075 DUAN Shihong, YU Tianqing, and HE Jie. WiDriver: Driver activity recognition system based on WiFi CSI[J]. International Journal of Wireless Information Networks, 2018, 25(2): 146–156. doi: 10.1007/s10776-018-0389-0 XIA Lu, CHEN C C, and AGGARWAL J K. Human detection using depth information by Kinect[C]. Computer Vision and Pattern Recognition 2011 WORKSHOPS, Colorado Springs, USA, 2011: 15–22. doi: 10.1109/CVPRW.2011.5981811. KOSBA A E, SAEED A, and YOUSSEF M. RASID: A robust WLAN device-free passive motion detection system[C]. 2012 IEEE International Conference on Pervasive Computing and Communications, Lugano, Switzerland, 2012: 180–189. doi: 10.1109/PerCom.2012.6199865. YANG Lei, LIN Qiongzheng, LI Xiangyang, et al. See through walls with COTS RFID system[C]. The 21st Annual International Conference on Mobile Computing and Networking, Paris, France, 2015: 487–499. doi: 10.1145/2789168.2790100. XIAO Jiang, WU Kaishun, Yi Youwen, et al. FIMD: Fine-grained device-free motion detection[C]. The 18th IEEE International Conference on Parallel and Distributed Systems, Singapore, 2012: 229–235. doi: 10.1109/ICPADS.2012.40. XI Wei, ZHAO Jizhong, LI Xiangyang, et al. Electronic frog eye: Counting crowd using WiFi[C]. IEEE INFOCOM 2014 - IEEE Conference on Computer Communications, Toronto, Canada, 2014: 361–369. doi: 10.1109/INFOCOM.2014.6847958. QIAN Kun, WU Chenshu, YANG Zheng, et al. PADS: Passive detection of moving targets with dynamic speed using PHY layer information[C]. The 20th IEEE International Conference on Parallel And Distributed Systems (ICPADS), Taipei, China, 2014: 183–190. doi: 10.1109/PADSW.2014.7097784. WU Chenshu, YANG Zheng, ZHOU Zimu, et al. Non-Invasive Detection of Moving and Stationary Human With WiFi[J]. IEEE Journal on Selected Areas in Communications, 2015, 33(11): 2329–2342. doi: 10.1109/JSAC.2015.2430294 ZHOU Zimu, YANG Zheng, WU Chenshu, et al. Omnidirectional coverage for device-free passive human detection[J]. IEEE Transactions on Parallel and Distributed Systems, 2014, 25(7): 1819–1829. doi: 10.1109/TPDS.2013.274 ADIB F and KATABI D. See through walls with WiFi![J]. ACM SIGCOMM Computer Communication Review, 2013, 43(4): 75–86. doi: 10.1145/2534169.2486039 DI DOMENICO S, DE SANCTIS M, CIANCA E, et al. WiFi-based through-the-wall presence detection of stationary and moving humans analyzing the Doppler spectrum[J]. IEEE Aerospace and Electronic Systems Magazine, 2018, 33(5/6): 14–19. doi: 10.1109/MAES.2018.170124 ZHU Hai, XIAO Fu, SUN Lijuan, et al. R-TTWD: Robust device-free through-the-wall detection of moving human with WiFi[J]. IEEE Journal on Selected Areas in Communications, 2017, 35(5): 1090–1103. doi: 10.1109/JSAC.2017.2679578 LI Fan, XU Cheng, LIU Yang, et al. Mo-sleep: Unobtrusive sleep and movement monitoring via Wi-Fi signal[C]. The 35th IEEE International Performance Computing and Communications Conference, Las Vegas, USA, 2016: 173–180. doi: 10.1109/PCCC.2016.7820634. 李姣軍, 余景鵬, 陶金, 等. 一維信號的小波去噪[J]. 重慶理工大學(xué)學(xué)報(bào): 自然科學(xué), 2016, 30(12): 83–89. doi: 10.3969/j.issn.1674-8425(z).2016.12.013LI Jiaojun, YU Jingpeng, TAO Jin, et al. Review of one-Dimensional signal wavelet de-noising[J]. Journal of Chongqing University of Technology:Natural Science, 2016, 30(12): 83–89. doi: 10.3969/j.issn.1674-8425(z).2016.12.013 LE CUN Y, BOSER B, DENKER J S, et al. Handwritten digit recognition with a back-propagation network[C]. The Advances in Neural Information Processing Systems 2, San Francisco, USA, 1990: 396–404. -
計(jì)量
- 文章訪問數(shù): 3154
- HTML全文瀏覽量: 1261
- PDF下載量: 134
- 被引次數(shù): 0
下載:
