Ricerc@Sapienza

The WiFi-STAR (WiFi based STand Alone Radar for indoor human activity monitoring) project aims to derive and propose innovative strategies for human activity monitoring in indoor environments by exploiting low-complexity and non-coherent radar techniques based on WiFi transmission for stand-alone sensors.
Thanks to the proliferation of WiFi devices and related applications, nowadays the availability of WiFi signals is very high in almost all private and public buildings such as houses, offices, stations, airports etc. Furthermore, there is an increasing demand for a capability to detect, accurately locate, and possibly recognize activities performed by a human using strategies that are devise-free and thus do not require any cooperation from the person being monitored nor use privacy-invasive technologies such as cameras. Moreover, in order to achieve a wide use of these sensors, it is important that each of them is small, light, low-cost and easy to install.
Based on these objectives and motivations, this project aims at tackling this issue, devising operative solutions that are based on WiFi radar technology but do not rely on any accurate synchronization between the transmitter and the receiver as well as any phase information extraction from the data. In order words, we aim at using effective non-coherent solutions that look for the presence of a target not relying on prior knowledge on the transmitted waveform but instead based on the amplitude modulation induced by the target motion on the received signal when observed across time.
The derivation and extensive validation of such strategies, against both simulated and real data collected in ad hoc acquisition campaigns, could pave the way for a distributed use of a WiFi-based network of stand-alone sensors. This could enable a complete monitoring of indoor environments, both for detecting the presence of unauthorized intruders and accurately localizing them but also for smart-home or e-healthcare applications.