Phd in INFORMATION AND COMMUNICATION TECHNOLOGY (ICT)

Thesis title: WiFi-based sensing: Signal Processing Techniques and Design Strategies

Through the years, target localization has captured the attention of both academic and industrial worlds, thanks to the huge amount of applications which require the knowledge of the position information. Several work can be found on this topic, where the target localization has been addressed in different ways, depending on the type of target, the specific application and the surrounding scenario. In this thesis approaches that relays on the WiFi Access Points (AP) as illuminator of opportunity, are taken in consideration. The main goal of this thesis is to propose low-cost and low-complexity techniques with the aim of introducing WiFi-based sensing technologies in everyday civilian application. The first approach relies on the use of WiFi-based Passive Radar (PR). Specifically, the conventional processing scheme is analyzed and several modification are proposed in order to reduce its overall complexity. Firstly, a rearrangement of the processing scheme is proposed to relax the data management. The Reciprocal Filter (RpF) is then considered as an alternative to the Matched Filter (MF) to operate the range compression stage. This is shown to provide several benefits for the overall processing scheme. Specifically, a simplified adaptive disturbance cancellation stage is devised, alongside to a unified processing scheme that doesn't depend on the adopted waveform modulation typically adopted in WiFi transmissions. Additionally, appropriate adjustments are also proposed to the theoretical RpF in order to cope with the inherent loss in terms of signal-to-noise ratio. The possibility of working without a reference channel is considered as a way to further reduce the architecture complexity. To this end, the processing is limited to the invariant initial portion of the physical layer protocol data unit, namely the PHY preamble, by synthetically building the reference signal according to the IEEE 802.11 Standard. These approaches are proven effective against simulated and experimental data. The use of a non-coherent approach is investigated as a way to devise a simplified sensor that doesn't rely on a reference channel (and signal), which results in a low complexity approach compared to the WiFi-based PR. Specifically, the presence of the moving target is determined by the amplitude modulation that induces on the direct signal from the WiFi Access Point (AP). The target discrimination capability of the sensor against the competing background is investigated and theoretically characterized. Hence, different solution, characterized by different complexity are proposed to deal with the background limitation. The analysis is validated against simulated and experimental data. Both the proposed techniques for target localization are characterized by a scarce range resolution, which implies that the target can not be positioned in a small environment (such as indoor scenarios) based on conventional radar approaches. To achieve an accurate positioning of the target inside the environment under observation, a mapping strategy that leverages other type of measurements, specifically Doppler and angle of arrival, from multiple receiving nodes is sought. The proposed strategy is theoretically characterized and tested for the effectiveness of the employed measurements and the impact of the network topology against simulated data.