Maintaining certain physical activity levels is important to prevent or delay the onset of many medical conditions such as diabetes, or mental health disorders. Traditional calorie estimation methods require wearing devices, such as pedometers, smart watches or smart bracelets, which continuously monitor user activity and estimate the energy expenditure. However, wearable devices may not be suitable for some patients due to the need for periodic maintenance, frequent recharging and having to wear it all the time. In this paper we investigate a feasibility of a devicefree human energy expenditure estimation based on RF-sensing, which recognises coarse-grained user activity, such as walking, standing, sitting or resting by monitoring the impact of a person's activity on ambient wireless links. The calorie estimation is then based on Metabolic Equivalent concept that expresses the energy cost of an activity as a multiple of a person's basal metabolic rate using Harrison-Benedict model. The experimental evaluation using low cost IEEE 802.15.4 transceivers demonstrated that the approach estimated energy expenditure within an indoor environment within 7.4% to 41.2% range when compared to a FitBit Blaze bracelet.

Detecting when a person leaves a room, or a house is essential to create a safe living environment for people suffering from dementia or other mental disorders. The approaches based on wearable devices, e.g. GPS bracelets may detect such events require periodic maintenance to recharge or replace batteries, and therefore may not be suitable for certain types of users. On the other hand, camera-based systems require illumination and raise potential privacy concerns. In this paper, we propose a device-free walking direction detection approach based on RF-sensing, which does not require a person to wear any equipment. The proposed approach monitors the signal strength fluctuations caused by the human body on ambient wireless links and analyses its spatial patterns using a convolutional neural network to identify the walking direction. The approach has been evaluated experimentally to achieve up to 98% classification accuracy depending on the environment.