Intelligent Alarm Technology Of Indoor Unexpected Fall

Chinese society gradually into the aging stage,the number of elderly population has more than 200 million.But China's pension-related industries are still in its infancy,the basic security needs of many elderly people are difficult to be met.Fall is a high frequency accident in the daily life of the elderly,but also leads to old people hurt and becomes one of the common causes of death.Existing human fall detection techniques can be divided into three categories: video-based detection technology,wearable sensing-based technology and environment-based sensing technology.Video detection technologies determine whether the elderly fall through the identification of image signals,but they are involved in privacy issues,and there may be monitoring dead ends and some wrong pattern recognition,so it is difficult to promote.Wearable sensingbased technologies are based on the principle of accelerating the sensor to detect.But many elderly people have a certain use of wearing equipment and resistance to psychological barriers.In addition,in the room where is easy to slip such as bathroom and toilet,the elderly will not use wearing equipment.The existing environment-based sensing technology relies mainly on audio signals to identify falls.The downside is that ambient noise can have a greater impact on the signal,and the sensors installed on the floor need to be powered by the battery,and it may make the elderly fall.It can be said that there is still a lack of a wide range of applications and solve the user's pain points fall detection technology.Based on the above thinking.In this paper,a new technology-optical fiber sensing technology is selected for the development of environmental sensing technology.The principle is based on Michelson interferometric fiber optic sensors: the fiber optic sensing system is installed under the floor,then the vibration of the floor and the disturbance of the environment can cause changes in the interference light signal.After collecting the data of the light wave,select the appropriate pattern recognition algorithm to identify the fall and other cases.Data collection experiment is done in three cases: in the absence of disturbance(no disturbance group),people walking on the floor(walking group)and the use of dummy to simulate the elderly to fall in different positions(fall group).Each group collected more than two thousand data.The experimental data is then displayed in Matlab.A two-dimensional eigenvector is constructed by two features(the number of times the signal passes through a certain threshold and the number of high and low numbers of the signal)after observing and analyzing signal.The eigenvector can distinguish the experimental data of the fall group from the other two groups well.At last,an algorithm for the eigenvector is written in Labview,and 100 experiments are carried out in three patterns.The results show that the accuracy of the algorithm is 100%,and its timeliness satisfies the requirement.