Experimental characterization of mobile fading channels aiming the design of non-wearable fall detection radio systems at 5.9 GHz

Abstract

One of the major concerns for the independent living of elderlies is a fall incident. To decrease human interaction errors and user privacy concerns of existing fall detection systems, a new generation of fall detection systems is emerging. The new trend is to design non-wearable devices that can monitor the physical activities of the home user using radio waves reflected off the body. This paper reports an in-home radio measurement campaign at 5.9 GHz, which has been conducted to study the impacts of different physical activities of the user, including fall incidents, on the channel transfer function (CTF) and the power delay profile (PDP) of indoor mobile radio channels. The home is equipped with a radio channel sounder that records the impulse response of the channel, in which the user behaves like a moving scatterer. Experimental results reveal that fingerprints of the user activities on the time-variant (TV) CTF and the TVPDP can be traced and classified both illustratively and computationally. It is shown that a so-called differential PDP (DPDP) ignores fine variations of the channel and highlights major changes, such as a fall incident. Mathematical expressions for the key characteristics of the channel are also derived under non-stationary propagation conditions. The results in this paper are important for the development of robust detection algorithms, which are an integral part of emerging fall detection systems using radio waves.