Research On Human Target Multi-dimensional Information Recognition Based On CSI

Wireless sensing technology based on channel state information(CSI)has broad application prospects in human-computer interaction,smart home and other fields due to its advantages of no target carrying equipment,no light and distance of view restrictions,and no privacy infringement.The existing studies have achieved satisfactory recognition accuracy for the human localization or activity information.However,many applications need to recognize not only the activity of the human body,but also the location of the human body.Only using commercial Wi Fi equipment for single target recognition cannot meet the development of the times and people's needs.Therefore,multi-dimensional information recognition of human targets has become an urgent problem to be solved.To solve the above problem,a multi-dimensional information recognition algorithm for human targets based on CSI decomposition(called CD-MDIR algorithm)is proposed.Specifically,we first decompose the CSI time series into Dynamic Location CSI(DLC)components which are affected only by human location and Dynamic Activity CSI(DAC)components which are affected only by human activity,according to the independent characteristics of the influence of human location and activity on CSI.Then,the Linear Discriminant Analysis(LDA)algorithm is used to map and transform the DLC components to enhance the location information.The activity information in the DAC components is enhanced by extracting the features of the DAC components.Finally,we designed a LSTM multi-dimensional information recognition network that successively identifies the location and activity information of the human body.Experimental results show that the proposed CD-MDIR algorithm achieves both higher location and activity recognition accuracy.