Runtime power management of energy harvesting real-time embedded system
Energy consumption has become the major roadblock to the progress of battery-powered systems, great interest has risen in powering these systems with renewable energy sources for overcoming the energy limitations. The conventional power management techniques are not designed for the Energy Harvestin...
Uloženo v:
| Hlavní autor: | |
|---|---|
| Médium: | Dissertation |
| Jazyk: | angličtina |
| Vydáno: |
ProQuest Dissertations & Theses
01.01.2012
|
| Témata: | |
| ISBN: | 1267651520, 9781267651525 |
| On-line přístup: | Získat plný text |
| Tagy: |
Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
|
| Shrnutí: | Energy consumption has become the major roadblock to the progress of battery-powered systems, great interest has risen in powering these systems with renewable energy sources for overcoming the energy limitations. The conventional power management techniques are not designed for the Energy Harvesting Real-time Embedded System (EH-RTES) and cannot handle the uncertainty in available energy. Therefore it is important to develop the novel power management techniques so that the EH-RTES is able to operate energy-efficiently and achieve energy autonomy. The goal of power management in this dissertation is to minimize deadline miss rate, which is a measure of the quality of service of the system, subject to the energy constraint. Task scheduling and DVFS are two main approaches used in power management of EH-RTES in this dissertation. We investigated and designed several new and novel power management methods and algorithms dedicated to energy harvesting systems and applications. One important investigation is to study the interaction between the optimization algorithms and the MPPT process and how to make them work better together with. With the new concept of effective energy dissipation, the adaptive optimization algorithms will be able to accurately capture the non-ideal characteristics of the system components. We also analyzed the complexity and performance of three real-time prediction models and evaluated their feasibility to energy harvesting speed prediction. As the multi-core architecture becomes more and more widely used in embedded systems, we proposed a low-complexity and effective task mapping, scheduling and power management method for multi-core real-time embedded systems with energy harvesting based on concept of task CPU utilization. The performance of the proposed power management and task scheduling algorithms are evaluated based on simulations. We have developed a discrete event-driven simulator in C++ and implemented the proposed algorithm with different applications. Comparing to the existing algorithms, the proposed algorithms achieve better performance in terms of the deadline miss rate and the system energy efficiency under various settings of workloads and harvested energy profiles. |
|---|---|
| Bibliografie: | SourceType-Dissertations & Theses-1 ObjectType-Dissertation/Thesis-1 content type line 12 |
| ISBN: | 1267651520 9781267651525 |