Project Group: Acoustic Indoor Localization

Tutors: Dr. Marcus Handte, Alexander Julian Golkowski

Pervasive Computing envisions context-aware systems and applications that adapt autonomously to the needs of their users. Knowledge about the location of persons and objects is often believed to be one of the most important prerequisites for context-aware adaptation. In outdoor settings, coarse grained location information can be gathered using satellite-based positioning systems such as GPS or Galileo. In indoor settings, WLAN- or BLE-based localization systems can achieve a more fine-grained accuracy in the range of meters. While this accuracy may be sufficient to implement applications in large or complex buildings such as airports, malls or storage facilities, most WLAN- and BLE-based systems are still too inaccurate for applications targeting smaller spaces such as offices or home environments.

The goal of this project group is to design and build several embedded systems that can send and receive audio signals to support localization tasks that require accuracies in the range of a few centimeters. Using these systems, the participants will develop application prototypes that demonstrate their benefits and limitations. The prototype applications can be freely designed by the participants and may range from simple acoustic beacons, e.g., to replace QR codes to 2D indoor localization with mobile phones, e.g., to support complex home automation scenarios.

From a technical perspective, the project encompasses embedded system design and construction as well as the application design, implementation, and validation. The participants will build models for PCBs and cases, solder hardware components and write software for microcontrollers and other application-dependent system components (e.g., mobile phones). For the hardware construction and application validation, the participants can use labs and offices of the Networked Embedded Systems group.

The admission to this course is managed centrally. If you have any questions, please contact marcus.handte@uni-due.de.