Prudent Automatic Falls Detection by Analyzing the Carrier State Information Signal Using Wi-Fi Devices

Abstract

Abstract This research article paves a way to rescue the elderly as soon as possible, thereby potentially saving the life of the person and reducing the costs borne by the sufferer. It has become very common for persons above the age of sixty to fall. An effective prevention and rescue operation is needed to ensure the life of older people. An automatic fall detection method is proposed by analyzing the channel state information (CSI) of the signal generated in the environment where the elderly live using commonly available Wi-Fi devices. The amplitude and phase of the CSI can be studied via Wi-Fi devices. The human activity is recognized based on the amplitude and phase difference of the receiving antenna. The features for actual fall and fall-like activities or pseudo fall are identified by considering the trivial power decline ratio (PDR) drafted in terms of the frequency and time domain of the waveform using a support vector machine. The performance of the proposed system is analyzed based on a confusion matrix for the false positive ratio and compared with a real time fall detection method designed using an ultrasonic array sensor.