University of Twente Student Theses

Login

An intelligent privacy-driven offloading algorithm for smart home applications

Rotmensen, Niels (2025) An intelligent privacy-driven offloading algorithm for smart home applications.

[img] PDF
867kB
Abstract:Smart home devices are becoming increasingly widespread. Some of these newer smart home applications require more processing power than is available on-device, and require external computational resources. While a common option is using cloud computing power, processing locally on the edge is a better solution in terms of user privacy as data does not leave the local network. Previous hybrid approaches either lack privacy considerations or do not consider the future resource availability when making offloading decisions. In this paper, we propose a hybrid approach that minimizes the privacy risk of offloading tasks to remote servers, considering future resource availability. To reach our goal, we built upon an existing privacy-enhanced offloading algorithm, with the aim of improving its performance by predicting future state, and scheduling tasks accordingly. These improvements give more flexibility in scheduling, where a privacy-sensitive task can wait in queue instead of immediately offloading. The results show that while the system load is high, ~1-1.5% fewer high privacy-sensitive tasks are offloaded compared to a baseline without prediction, while CPU utilization remains nearly identical for both approaches.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science BSc (56964)
Link to this item:https://purl.utwente.nl/essays/105146
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page