Training-Free Human Vitality Monitoring Using Commodity Wi-Fi Devices

Abstract

Device-free sensing using ubiquitous Wi-Fi signals has recently attracted lots of attention. Among the sensed information, two important basic contexts are (i) whether a target is still or not and (ii) where the target is located. Continuous monitoring of these contexts provides us with rich datasets to obtain important high-level semantics of the target such as living habits, physical conditions and emotions. However, even to obtain these two basic contexts, offline training and calibration are needed in traditional methods, limiting the real-life adoption of the proposed sensing systems. In this paper, using the commodity Wi-Fi infrastructure, we propose a training-free human vitality sensing platform, WiVit. It could capture these two contexts together with the target's movements speed information in real-time without any human effort in offline training or calibration. Based on our extensive experiments in three typical indoor environments, the precision of activity detection is higher than 98% and the area detection accuracy is close to 100%. Moreover, we implement a short-term activity recognition system on our platform to recognize 4 types of actions, and we can reach an average accuracy of 94.2%. We also take a feasibility study of monitoring long-term activities of daily living to show our platform's potential applications in practice.

A Systematic Review of Non-Contact Sensing for Developing a Platform to Contain COVID-19

“…Standing, walking, sleeping, eating, bathing and postures data are collected for developing a ML/DL model to recognize the fall, sleep disorder and diet of the patient. From the literature review, heart problems, fall, sleep disorder and eating habits can be recognized by non-contact wireless sensing platforms [ 14 , 15 , 17 , 20 , 22 , 23 , 25 , 26 , 27 , 28 , 30 , 31 , 33 , 34 , 36 , 38 , 40 , 45 , 47 , 48 , 49 , 57 , 61 , 62 , 63 ].…”

“…This platform can capture real-time human motion speed information without the offline training or calibration that involves human effort. The feasibility study of the platform revealed that it can monitor long-term activities of daily living in practice for various applications [ 36 ]. Eating is an essential activity in human daily life.…”

“…Wi-Fi sensing using commercial hardware is widely used because it is an inexpensive and easily available solution. Human activities have been monitored and classified in existing research by ML and DL algorithms having accuracies over 90% [ 28 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 ]. The average accuracy of promising non-contact technologies for monitoring human activities is shown in Figure 1 .…”

WiGId: Indoor Group Identification with CSI-Based Random Forest

“…By combining the amplitude information and phase information of CSI data, the robustness of the method is improved, and the performance of detection is also improved. Li [18] studied how to use commercial Wi-Fi equipment to detect people without a pre-training classification model, added speed detection as the basis of human activities, and achieved the highest detection accuracy of 100%. References [19][20][21][22] focused on the methods of detecting people's movements, discussed the influence of personnel movements on signal propagation, and studied how to identify people's movements with high accuracy in a changeable environment.…”

WiMorse: A Contactless Morse Code Text Input System Using Ambient WiFi Signals

“…In the past few years, WiFi signals have been exploited for localization [10], [11], [12], [13], [14], activity tracking [5], [6], [15], [16], [17], [18], [19], gesture recognition [20], [21], [22], [23] and recently more fine-grained respiration monitoring [24], [25], [23], [26] and material sensing [27]. Even though non-periodical fine-grained activities such as keystrokes, finger gestures and mouth speaking have been reported to be sensed using WiFi Channel State Information (CSI), they all rely on one assumption that the position of the moving target relative to the WiFi transceivers is fixed such that the same movement would lead to similar signal change patterns in different rounds.…”