Low-power fall detection in home-based environments

Abstract

Fall detection of the elderly becomes more critical in an aging society. However, how to put forward fall detection with reliability and high accuracy while maintaining real-time and energy-efficiency is an important issue. To this end, we design and implement an energy-efficient prototype called Asgard, in which a fall detection algorithm and a hybrid energy-efficient strategy are proposed. The algorithm, which can flexibly track the body change by recovery angle detection, helps to reduce the false positive phenomenon as well as detection time (DT). Results of comprehensive evaluations show the accuracy rate of 96.25%, which is higher than AMD (Advanced Magnitude Detection). More notably, the prototype still has low DT with the aforementioned accuracy. More precisely, with the proposed hybrid energy-efficient algorithm, Asgard functions well for approximately one month using only two AA batteries (1500mAH each).

Accurate Fall Detection and Localization for Elderly People Based on Neural Network and Energy-Efficient Wireless Sensor Network

“…The changing of patient location from one place to another can provide a future opportunity to continue the current study to further understand and capture the relevant challenges of the patient's location, especially in practical applications. [68],fall detection [115],human activities [87],fall detection [116],cardiovascular and respiratory [50],fall detection [90],fall detection [117],human activities [118],fall detection [119],human activities [24],fall detection [71],patient temperature [120],human activities [121],elderly living Proposed S-BFDA Estimated battery life (h) Figure 27. Comparison between battery life of the proposed S-BFDA and previous system.…”

“…The previous works are plotted on the x-axis, and the estimated battery life of the sensor node is plotted on the y-axis, as shown in Figure 27. For a fair comparison between the proposed FDS and previous works [24,50,68,71,87,90,[115][116][117][118][119][120][121], the battery life of the proposed systems in the previous works was recalculated based on 1000 mAh. Most of these works are similar to our proposed system because they use different fall detection algorithms to monitor elderly fall events or activities.…”

Chameleon: personalised and adaptive fall detection of elderly people in home-based environments

“…Asgard is implemented to detect a fall, the prototype of which is constructed as shown on the top of the figure, while the function components are shown at the bottom. Asgard is a smart accelerometer sensor consisting of a MMA7260Q triaxial accelerometer, a microcontroller (abbreviated as MCU), and CC2520 radio model (a Zigbee module) (Ren et al, 2012). It has been implemented in our former research and is chosen as a data collection and fall detection evaluation platform.…”

“…It also shows the impacts of fall severe (FallS as shown in Figure 1) and fall light (FallL as shown in Figure 1) activities from 6 young group people aged from 19 to 27, and the impact of 95% upper confidence interval (Fall UCI as shown in Figure 1) of FallS and FallL. The data is collected when the volunteers do the ADLs continuously or simulate designed fall action wearing Asgard (Ren et al, 2012). Asgard is a smart sensor, constructed and used to collect the acceleration impacts of various activities at a sampling rate of 62.5 HZ with sensitivity of 200 mV/g.…”

“…In contrast, if the value is too low, the system can detect all actual falls successfully, but, at the same time, it may generate some false alarms. However, most studies have determined this threshold using empirical data (Bourke et al, 2006;Kangas et al, 2008;Ren et al, 2012) extracted from the impact data of young people. Evidently, this empirical threshold cannot be used directly by the elderly people, so we argue that a fixed threshold is not flexible enough to detect falls for different people.…”