Detecting from in-vehicle at the physical layer a smart device spatial location via 2.4 GHz SRD RF services. Innovation report

Preview

PDF WRAP_Theses_Marsic_2022.pdf - Submitted Version - Requires a PDF viewer. Download (3MB) | Preview

Request Changes to record.

Abstract

Detecting the position of a smartphone over a short range outside and inside a vehicle through its secondary radio frequency (RF) services is a difficult task, with various trials and tests employing Wi-Fi, Bluetooth Low Energy (BLE), and Near Field Communication (NFC) suggesting possible solutions using forthcoming services to be introduced, such as the ultra-wide band (UWB) radio or the IEEE 802.15.4 standard. This thesis addresses the modern radio location problem of inside-outside position location for a mobile RF source by proposing an empirical solution that is based on the 2.4 GHz IEEE 802.15.4 standard. For a proof-of-concept scenario involving a static vehicle with its side windows down and no passenger presence, this work derived an offline calculation based on empirical measurements demonstrating that four directive, inside-orientated sensors can be sufficient to produce a 100% successful inside-outside RF position discrimination. Furthermore, by adding four more RF directive sensors pointing towards outside, it was possible to detect on a radius of up to 5 m around the vehicle the variable location of a mobile RF source within a root mean square error (RMSE) of up to ±1.4 m outside and ±0.4 m inside the vehicle. The demonstrated experimental RF ranging is based solely on the received signal strength indicator (RSSI) and the individual sensor’s directivity achieved through shielding. The results rely on an RSSI fingerprinting database derived from only empirical outdoor measurements, delivering a consistent performance inside the highly RF-reflective vehicle cabin by exploiting the sensor position and directivity and suggesting that a focus on the front of each seat may avoid future human interference. Moreover, a theoretical propagation model based on Friis’ transmission equation constructed on system parameters shows a high correlation with the RSSI fingerprinting experimental model, thus supporting the consistency of the empirical model. The directive RF sensors used in the experimental work were refined based on ray tracing (RT) simulation suggestions so as to display an optimal discrimination of targets on a convex open volume such as a vehicle cabin.

To summarise, this EngD successfully demonstrated a 2.4 GHz RF location system based on the IEEE 802.15.4 standard with direct applications in RF detection, such as the restricted entrance scenario employed in vehicle and smart infrastructure industry.

Request changes or add full text files to a record

Repository staff actions (login required)

View Item