Wireless Hand Gesture Recognition Based On Continuous-wave Doppler Radar Sensors

Continuous-wave Doppler radar sensor was often used in non-contact motion detection in the past.In this dissertation,we try to improve and apply this technology into hand gesture recognition(HGR),which is a novel and challenging issue.On the aspect of hardware,this dissertation uses direct frequency conversion architecture,subcarrier modulation and band-pass sampling to solve the problem of baseband DC offset and low frequency flicker noise in traditional radar architecture,and greatly reduce the hardware cost.On the aspect of algorithm,this paper proposes an Arcsine high precision phase demodulation algorithm which is more suitable for motion detection,and a tracking mapping algorithm for multiple channel receiver data fusion.Through the designed hardware,this paper validates the validity of the proposed two algorithms in various test environments,and achieves the purpose of linear motion recovery,two-dimensional human trajectory tracking and motion trajectory mapping.Using the tracking mapping algorithm proposed in this paper,it is theoretically possible to demodulate and track the trajectory of a human object in a two-dimensional plane or three-dimensional space.The proposed mapping algorithm can map the complex motion of human hands to linear space,which can help the followed processing of behavior pattern recognition.Experiments show that the hardware and algorithm proposed in this paper can be applied to the field of hand gesture detection very well.