Orientation-Independent Human Activity Recognition Using Complementary Radio Frequency Sensing.

RF sensing offers an unobtrusive, user-friendly, and privacy-preserving method for detecting accidental falls and recognizing human activities. Contemporary RF-based HAR systems generally employ a single monostatic radar to recognize human activities. However, a single monostatic radar cannot detect the motion of a target, e.g., a moving person, orthogonal to the boresight axis of the radar. Owing to this inherent physical limitation, a single monostatic radar fails to efficiently recognize orientation-independent human activities. In this work, we present a complementary RF sensing approach that overcomes the limitation of existing single monostatic radar-based HAR systems to robustly recognize orientation-independent human activities and falls. Our approach used a distributed mmWave MIMO radar system that was set up as two separate monostatic radars placed orthogonal to each other in an indoor environment. These two radars illuminated the moving person from two different aspect angles and consequently produced two time-variant micro-Doppler signatures. We first computed the mean Doppler shifts (MDSs) from the micro-Doppler signatures and then extracted statistical and time- and frequency-domain features. We adopted feature-level fusion techniques to fuse the extracted features and a support vector machine to classify orientation-independent human activities. To evaluate our approach, we used an orientation-independent human activity dataset, which was collected from six volunteers. The dataset consisted of more than 1350 activity trials of five different activities that were performed in different orientations. The proposed complementary RF sensing approach achieved an overall classification accuracy ranging from 98.31 to 98.54%. It overcame the inherent limitations of a conventional single monostatic radar-based HAR and outperformed it by 6%.

Time-Frequency Characteristics of In-Home Radio Channels Influenced by Activities of the Home Occupant.

While aging is a serious global concern, in-home healthcare monitoring solutions are limited to context-aware systems and wearable sensors, which may easily be forgotten or ignored for privacy and comfort reasons. An emerging non-wearable fall detection approach is based on processing radio waves reflected off the body, who has no active interaction with the system. This paper reports on an indoor radio channel measurement campaign at 5.9 GHz, which has been conducted to study the impact of fall incidents and some daily life activities on the temporal and spectral properties of the indoor channel under both line-of-sight (LOS) and obstructed-LOS (OLOS) propagation conditions. The time-frequency characteristic of the channel has been thoroughly investigated by spectrogram analysis. Studying the instantaneous Doppler characteristics shows that the Doppler spread ignores small variations of the channel (especially under OLOS conditions), but highlights coarse ones caused by falls. The channel properties studied in this paper can be considered to be new useful metrics for the design of reliable fall detection algorithms. We share all measured data files with the community through Code Ocean. The data can be used for validating a new class of channel models aiming at the design of smart activity recognition systems via a software-based approach.