An Experimental Platform for Fall Detection Using Beacon, Node MCU and MATLAB

Abstract

Healthcare wearables have become very powerful and useful devices capable to detect and monitor numerous health parameters and physical conditions. Fall Detection Systems (FDS) are a part of them, with function to detect the falls, mainly in adults. Real-time falls detection may reduce the risk of major problems and enable protective and medical personnel to act immediately. FDS varies in terms of embedded hardware and software (algorithms). Here, we present a system that can be used as an experimental platform to explore the fall detection algorithms based on inertial methods. Battery powered Bluetooth Low Energy (BLE) Beacon with built in accelerometer is used as a sensor device and data transmitter. BLE ESP based gateway receive the data and forward them to MATLAB host, on which experiments are conducted to find the most suitable algorithms. Later, the algorithms are embedded into suitable platforms, which are incorporated into sensor housings, receiver-indicators or home/hospital care systems. The architecture of overall experimentations system as well as preliminary testing results for accelerometer-based algorithms are presented, but without changing the hardware configuration, other algorithms can also be tested.