Loading...

Energy- and Reliability-Aware Active Strategies for Mixed-Criticality Cyber-Physical Systems

Toghani Roshan, Mohammad Mehrab | 2025

0 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 57993 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Ansari, Mohsen; Hessabi, Shaahin
  7. Abstract:
  8. The expansion of Cyber-Physical Systems (CPSs) in diverse domains, including automotive, aerospace, and healthcare, has led to increased challenges for these systems in the last decade. With the increasing complexity of the functions performed by these systems, Mixed-Criticality Systems (MCSs) that integrate tasks with different criticality levels have attracted much attention. These systems often have a heterogeneous multiprocessor architecture and use Directed Acyclic Graphs (DAGs) to model task dependencies. This paper proposes a fault-tolerant scheduling algorithm that manages energy consumption and reliability while considering the Deadline Miss Ratio (DMR) in heterogeneous multiprocessor systems. Our proposed scheduler uses Dynamic Voltage and Frequency Scaling (DVFS) and Dynamic Power Management (DPM) techniques to reduce energy consumption. It also enhances system reliability through joint N-Version Programming (NVP) and task replication fault-tolerant techniques. Our experimental results show that our approach achieves 24.5\% (on average) energy savings compared to the state-of-the-art methods and significantly improves system reliability while maintaining the DMR
  9. Keywords:
  10. Mixed-Criticality Emdedded Systems ; Cyber-Physical Systems ; Energy Consumption ; Reliability ; Reliability Aware Energy Management ; Energy Management ; Directed Acyclic Graph (DAG)

 Digital Object List

 Bookmark

...see more