Fall Detection Algorithm of the Elderly Based on BP Neural Network

With the rapid development of population aging, falling is a very serious problem for the elderly. Real-time detection of whether the elderly has fallen can minimize the damage caused by falling. Therefore, this paper proposes a fall detection method based on BP neural network. The algorithm uses a three-layer BP reverse neural network to collect human motion data by wearing a three-axis acceleration sensor (MMA7660FC). After feature extraction of the data, network training is carried out, and the neuron weight and learning rate are adjusted to the training process so that it can realize the function of fall detection. Experimental results show that the algorithm can identify falls well, and its accuracy rate can reach 99.44%.