TY - JOUR
T1 - Blind Separation of Doppler Human Gesture Signals Based on Continuous-Wave Radar Sensors
AU - Gu, Zhitao
AU - Wang, Jun
AU - Shen, Fazhong
AU - Xu, Kuiwen
AU - Ye, Dexin
AU - Huangfu, Jiangtao
AU - Li, Changzhi
AU - Ran, Lixin
N1 - Funding Information:
Manuscript received December 18, 2018; accepted January 17, 2019. Date of publication February 26, 2019; date of current version June 7, 2019. This work was supported by the NSFC under Grant 61771421, Grant 51607168, Grant 61471315, Grant 61601161, and Grant 61771422. The Associate Editor coordinating the review process was Christoph Baer. (Corresponding author: Lixin Ran.) Z. Gu, F. Shen, K. Xu, D. Ye, J. Huangfu, and L. Ran are with the Laboratory of Applied Research on Electromagnetics, Zhejiang University, Hangzhou 310027, China (e-mail: ranlx@zju.edu.cn).
Publisher Copyright:
© 1963-2012 IEEE.
PY - 2019/7
Y1 - 2019/7
N2 - Recently, progresses have been made in hand gesture recognition based on Doppler radar sensors. However, it remains a technical challenge to avoid the interfering human motions. An example is detecting hand gestures in the presence of random body movements. In this paper, we propose a solution based on a Single-input Multiple-output frontend and a blind motion separation algorithm. Assisted by an additional receiving channel, Doppler signals caused by different motions can be separated by extending the algorithm originally developed for separating human voices. The experimental separation of hand gestures from interfering movements verified the effectiveness of this approach. The proposed solution can be potentially used in novel applications such as human gait and gesture recognitions.
AB - Recently, progresses have been made in hand gesture recognition based on Doppler radar sensors. However, it remains a technical challenge to avoid the interfering human motions. An example is detecting hand gestures in the presence of random body movements. In this paper, we propose a solution based on a Single-input Multiple-output frontend and a blind motion separation algorithm. Assisted by an additional receiving channel, Doppler signals caused by different motions can be separated by extending the algorithm originally developed for separating human voices. The experimental separation of hand gestures from interfering movements verified the effectiveness of this approach. The proposed solution can be potentially used in novel applications such as human gait and gesture recognitions.
KW - Blind motion separation
KW - Doppler radar sensor (DRS)
KW - gesture recognition
KW - independent component analysis (ICA)
UR - http://www.scopus.com/inward/record.url?scp=85067128943&partnerID=8YFLogxK
U2 - 10.1109/TIM.2019.2896364
DO - 10.1109/TIM.2019.2896364
M3 - Article
AN - SCOPUS:85067128943
SN - 0018-9456
VL - 68
SP - 2659
EP - 2661
JO - IEEE Transactions on Instrumentation and Measurement
JF - IEEE Transactions on Instrumentation and Measurement
IS - 7
M1 - 8653471
ER -