Hand movement classification from measured scattering parameters using deep convolutional neural network

• Published in: Measurement : journal of the International Measurement Confederation Vol. 151; p. 107258
• Main Authors: Gupta, Sindhu Hak, Sharma, Aayush, Mohta, Mohit, Rajawat, Asmita
• Format: Journal Article
• Language: English
• Published: London Elsevier Ltd 01.02.2020; Elsevier Science Ltd
• Subjects: Accuracy; Antennas; Artificial neural networks; Classification; Compact dual-band antenna; Computer simulation; DCNN; Human body; Human motion; Human movement; Motor ability; Movement; Network analysers; Neurodegenerative disorders; Parameter identification; Rehabilitation; S11; S21; Sensors; UWB; WBAN; DCNN; WBAN; UWB; Accuracy; S11; S21; Compact dual-band antenna; Human movement; Neurodegenerative disorders
• ISSN: 0263-2241; 1873-412X
• DOI: 10.1016/j.measurement.2019.107258

•Electromagnetic signals radiated by antenna during different movements are measured by VNA.•For DCNN based classification, frequency domain analysis has been carried out.•Data set has been recorded with the help of fabricated transmitting and receiving antennas.•UWB range has been chosen as it is safe for human body communication.•No additional sensors such as accelerometers and gyroscopes are required for evaluation. Human body movement analysis aids in implementing the physical rehabilitation process to regain the diminished motor abilities. In this work, the feasibility of using antennas and no dedicated sensors for movement identification is explored. Compact dual-band transmitting and receiving antennas of size 37.6 mm × 27 mm with frequency accuracy of 87% at lower band and 76% at higher band are simulated, fabricated and placed on the body of ten healthy subjects with normal BMI (18.5–24.9) kg/m2. Subjects are made to demonstrate five different hand movements. The dataset for each hand movement is experimentally measured using a Vector Network Analyzer (VNA). Measurement results reveal that the Reflection and Transmission coefficients (S11 and S21) of on-body antennas for each hand movement exhibit unique channel functionalities with respect to frequency. The uniqueness of the exhibited parameters aids in identifying the hand movements. Classification of hand movements based on measured data set is carried out using Deep Convolutional Neural Network (DCNN). The classification accuracy of movement comes out to be 93.32% when classifying using S11 parameters, and an accuracy of 98.67% when classifying using S21 parameters.