Fall-Attention: An Attention-Based Fall Detection Method for Adjoint Activities

WiFi-based wireless sensing has gained popularity for enabling smart indoor services, one of which is fall detection which plays a vital role in mitigating health risks for elders. Previous approaches have treated daily activities as independent events and built models to distinguish falls from others. However, human activities are usually adjoint in practice, e.g., the elder may suddenly fall when she walks. This adjoining introduces shared features between different activities, thereby affecting the classification performance. To address this problem, we propose Fall-attention, an attention-based fall detection method that can focus on the features related to fall events and suppress interference of irrelevant activities to improve performance. Its basic idea is to produce a task-oriented feature representation of fall events inside the signal using attention-based sentence embedding techniques and Recurrent Neural Network (RNN). We incorporate multi-task learning into Fall-attention by adopting multiple independent classification modules. This enables the model to explore different regions of the signal, capturing the composition of adjoint activities. A series of signal preprocessing and data enhancement techniques are also adopted to promote model training. Experimental results of the dataset containing adjoint activities demonstrate the superiority of Fall-attention over previous methods, which achieves an average accuracy of 95%.

I. Introduction

Falls, causing around 684,000 annual global fatalities, primarily affecting those aged 60 and above, create a pressing issue [1]. With an aging population adding further strain to elder healthcare systems, the development of automated indoor fall detection systems has become critically essential. Researchers have pursued this goal by employing various sensing methods, including wearable devices [2], computer vision (CV) [3], [4], and wireless sensing [5], [6], [7], [8], [9], [10]. While wearable devices equipped with inertial sensors can measure user velocity, they may cause discomfort among the elderly [11]. CV-based methods offer non-contact fall detection through video monitoring but grapple with challenges related to camera obstruction and privacy. In contrast, wireless sensing, especially WiFi-based, is cost-effective and non-intrusive, utilizing ambient wireless signals reflected by the human body to gather activity data without disturbing the user [12]. This approach, leveraging existing infrastructure like public access points (APs), significantly reduces deployment costs, leading to numerous fall detection applications [11], [13], [14], [15], [16].