Research On The Correction Technology Of Jump Phenomenon In The Field Of WiFi Human Behavior Recognition

With the rapid development of wireless technology and artificial intelligence,human behavior recognition based on WiFi signals has attracted more and more attention from the academic community.Common human behavior recognition based on computer vision and special sensors is difficult to be widely applied due to the disadvantages of being vulnerable to environment and obstacles,high cost,and poor user experience.In this paper,the jump phenomenon is found through experiments,and it is demonstrated that jump phenomenon affects the eigenvalues used for behavior classification and leads to inaccurate human behavior recognition.It is of great practical significance and wide application prospect to study the problem of correcting the jump phenomenon in the field of WiFi human behavior recognition.This paper expounds the principle of human behavior recognition based on WiFi signal,which lays a theoretical foundation for the research of this paper.This paper introduces the research background and current situation of WiFi human behavior recognition and discusses four representative WiFi human behavior recognition systems in detail.Aiming at the jump phenomenon,the reason of jump phenomenon is explained theoretically at first,and then the algorithm to correct jump phenomenon is put forward.The main research work is as follows:(1)According to the characteristics of received data and the theoretical knowledge of signal propagation,the jump points are identified by calculating the Pearson correlation coefficient between adjacent CSI data frames.By proposing relevant hypotheses and studying the problem of correcting jump phenomenon in specific situations,the problem is reduced to an optimization problem.The nested Switch-Mode firefly algorithm is proposed to solve the optimal initial replacement position,so that the average value of similarity between each data segment is the largest after the corrected data is segmented.Among them,the SimCsiData model is proposed to calculate the similarity of two CSI data segments.(2)The performance of the proposed algorithm is evaluated in three experimental scenarios,and the effect of the proposed algorithm on the correction of jump phenomenon is demonstrated from different aspects.The three experimental scenarios are: line-of-sight static environment,non-line-of-sight static environment,and waving at a fixed frequency in the case of line-of-sight.The experimental results show that the nested Switch-Mode firefly algorithm significantly corrects jump phenomenon by combining SimCsiData model to calculate the similarity between two CSI data segments.