Research On Vital Sign Detection Method Based On Sparse Array MIMO Rada

The non-contact vital sign detection technology can detect various parameters without contacting the human body,such as breath detection,heartbeat detection,etc.the application scenarios are also extended to many aspects,such as disaster relief scenario,disease detection,sleep scenario detection,etc.In many application scenarios,multi person vital signs detection needs to be carried out in the whole scene.At present,the commonly used multi person separation algorithm is digital beam forming(DBF).However,when the distance between two targets is too close to complete the separation,the angular resolution needs to be improved.The first method can increase the number of antenna elements,but it will increase the hardware pressure and improve the system complexity;The second method can use sparse array.However,the existing sparse array algorithm for vital sign detection direction is suitable for single angle beam pointing,which is difficult to cover all the angle range of the actual scene.To solve these problems,a MIMO radar system based on sparse array is designed to detect the vital signs of multiple people in the scene.The main work and research contents are summarized as follows:(1)The basic principle of MIMO Radar Based on continuous wave(CW)and the principle of multi person vital signs detection based on DBF are introduced.According to the actual detection requirements,a multiple input multiple output(MIMO)radar system is built.Experiments show that the angle information of the target can be accurately obtained by DBF after amplitude and phase correction.(2)This paper introduces the pattern calculation method based on two-dimensional sparse array,and then introduces two classical one-dimensional sparse array algorithms:matrix pencil method(MPM)and particle swarm optimization(PSO).Taking the sleep scene as the actual scene,this paper proposes a combination algorithm of particle swarm optimization and convex optimization: determine the sparsity according to the actual scene,calculate the angle constraint range,and use the convex optimization algorithm to obtain the complex excitation coefficient instead of the guidance vector;The peak sidelobe levels of the pattern when the beam points to all angles in the angle range are compared,and the optimal sparse array is obtained by iterative solution.The simulation results show that the optimal sparse array not only improves the angular resolution,but also keeps the peak sidelobe level stable when the beam direction changes.(3)The principle of vital signs detection based on MIMO radar is introduced.An iterative circular curve fitting algorithm is proposed to calculate the DC offset,remove the influence of DC on the phase extraction process,and extract the phase information by difference and cross multiplication.The experimental results show that the iterative circular curve fitting method can better estimate the DC offset and extract more accurate phase information than the traditional fitting algorithm.(4)According to the optimal sparse array,a MIMO radar experimental system is built,and four vital signs detection experiments are carried out: single field center,single field edge,double forward lying down and double reverse lying down.The results show that within the two-dimensional angle range required by the experimental scene,the system in this paper can accurately detect single and multiple human vital signs signals.In each experiment,the correlation coefficient between the experimenter’s heartbeat time-frequency diagram curve and the corresponding reference signal is more than 0.86,and the average relative error is less than 0.006.Human experimental results verify the effectiveness of the system and algorithm in this paper.