Abstract:
In this project we presents a two-dimensional phase extraction system using
passive WiFi sensing, coupled with end-to-end deep learning framework to moni-
tor three basic elderly care activities namely: breathing rate, essential tremor and
falls. Speci cally, a WiFi signal is acquired through two channels where the rst
channel is the reference one, whereas the other signal is acquired by a passive re-
ceiver after re
ection from the human target. Adaptive lter is performed to make
the surveillance signal source-data invariant by eliminating the echoes of the direct
transmitted signal. We propose a novel convolutional neural network to classify the
complex time series data and determine if it corresponds a to any of three mentioned
activities, followed by a two dimensional phase extraction platform to determine and
track the activity, lastly, benchmarking with random forest estimator is also per-
formed. We collect an extensive dataset to train the learning models and develop
reference benchmarks for the future studies in the eld. Using signal processing
of cross-ambiguity function, various features in the signal are extracted. The en-
tire implementations are performed using software de ned radios having directional
antennas. We report the accuracy of our system in di erent conditions and envi-
ronments and show that breathing rate can be measured with an accuracy of 87%
when there are no obstacles. We also show a 98% accuracy in detecting falls and
93% accuracy in classifying tremor. The results indicate that passive WiFi systems
coupled with deep learning show a great promise in replacing typical invasive health
devices as standard tools for health care.
