Research On Motion Recognition,Multi-Target Recognition And Indoor Positioning Algorithm Based On Multi-Sensor

With the continuous progressing of informational epoch and improving rapidly of electronic technique,contactless sensor represented by radar attracts widespread attentions.We can detect targets and obtain their location,motion direction and speed by analyzing electromagnetic wave received by radar.Moreover,with the specific radar signal processing algorithm,we can apply radar to a wider range of application,such as human motion recognition,multi-target recognition and indoor positioning,etc.In this paper,the radar signal processing algorithm is studied and the main research contents and innovations are as follows:(1)A multi-classification algorithm for human motion recognition based on Impulse-Radio Ultra-wideband(IR-UWB)radar is presented.The algorithm includes three parts.First,the k-NearestNeighbor(kNN)algorithm is used to classify the radial features of pre-processed signal to determine the subject's radial displacement direction.Then,the power spectrum feature extraction algorithm and Doppler shifts feature extraction algorithm are proposed to extract and visualize the characteristics from the different categories classified by the first part.Finally,the feature spectrograms obtained by the second part are sent into Convolutional Neural Networks(CNNs)for training and testing to realize the recognition of human motions.To verify the performance of proposed algorithm,dataset was created from 15 persons including 12 kinds of motions.The Five-Fold Cross Validation was conducted to calculate the recognition accuracy.As a result,the average accuracy of judging the radial displacement directions of subjects was up to 99%.Furthermore,the average accuracy of estimating the motions of subjects reached 98%.Experiments have proved that the proposed algorithm can achieve high recognition accuracy in daily human motions and is feasible in a variety of test environments.(2)A integrated signal pattern algorithm for multi-target recognition based on Impulse-Radio Ultra-wideband(IR-UWB)radar is presented.The algorithm includes three parts.First,the original radar echo signal is preprocessed to remove the background and spatial clutter,and the range compensation is carried out.Then,the signal pattern features are extracted by cluster feature extraction method based on UWB channel model.Finally,the signal pattern features are classified by k-NearestNeighbor(kNN)algorithm.To verify the performance of proposed algorithm,the dataset containing 0-5 people was created.The Five-Fold Cross Validation was conducted to calculate the recognition accuracy.As a result,the average accuracy of the algorithm is more than 90%in the open indoor environment.Experiments show that the proposed algorithm has low computational complexity and is feasible to detect the number of people of the region in a variety of test environments.(3)An indoor localization algorithm based on millimeter wave frequency modulated continuous wave(FMCW)radar is presented.The algorithm includes three parts.First,the background noise is removed by using background spectrum subtraction method and moving target indication(MTI)algorithm.Then,the target range,velocity and angle information are detected by using range FFT,velocity FFT and angle FFT respectively.Finally,the obtained range,velocity and angle information are integrated and identified in the azimuth map.To verify the performance of proposed algorithm,the measurements tools were used to measure the real range and angle information,and the results were compared with the radar detection results.As a result,the average range error is less than 10cm and the angle error is less than 5° in the open indoor environment.