The Research Of Person Status Recognition Based On Wireless Sensing

The Research of Person Status Recognition Based on Wireless Sensing belongs to a kind of emerging technology,which is a further advance of passive wireless localization and activity recognition.And its theory is that human has effected on wireless signals in wireless net,and the influence will be different due to the difference of person status.Thus,with analysis of received signals we could realize person status recognition.Now,our lives are filled with wireless signals,such as FM,WiFi and so on.Thus,the wireless signals in the space provide basis of researching and apply.Meanwhile,it makes wireless sensing economy.The passive wireless sensing need not carry any equipment with wireless signals,compared with traditional technique which is convenient for practical application.The technique has broad application prospects in smart home and security aid.The key point of research on wireless sensing in human identification is that recognizing the influence of wireless signals affected by the target.Wireless sensing technique can be divided into two categories,based on model and based on pattern recognition sensing technique.The key of the former is building the model and the key of the latter is extracting features.The paper explores the wireless sensing in the two categories respectively and certify the advantages of the proposed method with experience.The paper proposes a dynamic differential phase analyze method for wireless sensing at building model.The model proposes utilizing dynamic signals that play a positive role in sensing system to realize wireless sensing.And utilize the rate of dynamic differential phase to realize sensing.The model utilizes the change rate of dynamic differential phase to analyze the influence of wireless signals in space,so as to choose the better sensing region.Besides we propose a scale-variable receiving antennas distance scheme,which can design the optimal antenna placement under the premise of fixed the target.Then the proposed model can guide the experimental design.In based on pattern recognition sensing technique the paper proposes based on respiration human identification and utilize universal and robust intrinsic features to realize identification.Compared with traditional technique proposed system utilizes unique respiration to realize high accuracy identification,which is universal.In addition,the intrinsic features can largely solve the problem that traditional features are sensitive to environmental noise,and sensing performance is stable.To certify feasibility and advantages of above two methods,the paper sets a mass of experience.With letter recognition,respiratory monitoring,and AOA estimation the dynamic differential phase model can be certified to improve sensing performance.With changing environment and compared experience based on intrinsic features human identification can be proved to be universal and achieve higher accuracy.