Blind Separation Of Multiple Motions Based On Doppler Radar Sensing

In recent years,the desired for physical health and home security has increased rapidly.Thus more attention has been paid to the field of vital signal monitoring and multiple motion detection.However,the current detection method based on image sensors was limited seriously in a complex environment with many obstacles.With the rapid development of integrated radio frequency circuits(RFICs),miniaturized continuous-wave(CW)Doppler radar sensors(DRSs)are used more and more widely in wireless detection of human vital signs and behaviors,which have advantages in simple structure,low power consumption and strong penetrability.Even so,in many previous researches,the detection of multiple motions always suffers from the problem of mutual interference.It remains a technical challenge for a DRS to linearly separate independent motions without any prior infornation.Based on Doppler radar sensing technology,this dissertation covers a series of researches on the DRS architecture and blind separation problem of multiple moving targets.Firstly,the theoretical model of inverse problem is established based on blind signal separation(BSS)principles.Secondly,low-IF single-input multiple-output(SIMO)DRS system and distributed multiple-input multipfe-output(MIMO)micro-doppler sensor are designed and manufactured for motion separation.Then,the DRS system detects the signals scattered off multiple objects,and a two-step blind motion separation approach is proposed to recover the 1-D motions.Finally,combining with the correlation coefficient matching algorithm and a two-dimensional positioning algorithm,the two-step method is extended to separate 2-D motions.The experimental results and error analysis show that the SIMO DRS and the proposed motion separation approach can successfully separate combinations of 1-D motions,such as triangular,sinusoidal,random motions and hand gestures from interfering movements,as long as the veloeities or initial phases of such motions are different.With the two-step approach sequentially tackling the signal separation and the nonlinear demodulation problems,both small-and large-scale independent motions can be linearly separated.In the meanwhile,preliminary research is carried out on the separation of 2-D motions to verify the effectiveness of our approach.The above researches imply its tremendous potential in tracking and localization of multiple moving objects in indoor environments.