Date of Award

4-1997

Degree Name

Master of Science in Engineering

Department

Electrical and Computer Engineering

First Advisor

Dr. Sharon Hu

Second Advisor

Dr. Garrison Greenwood

Third Advisor

Dr. Frank Severance

Access Setting

Masters Thesis-Open Access

Abstract

This thesis aims at developing efficient analytical techniques to perform trade-off studies in the architectural design exploration phase of real-time embedded systems. Real-time embedded systems (RTES) have stringent timing requirements. In the process of designing such a system, a key issue is to determine if a system is able to meet all the timing requirements imposed on it, though it may prove attractive in terms of cost and performance. Traditional methods of using event-driven simulation for modern RTES are very time consuming and are not guaranteed to prove feasibility. Specifically, we identify a flexibility metric to compare various options in the design exploration task. This metric is obtained based on certain schedulability bounds. The famous rate monotonic scheduling theory was studied in detail. The new metric was evaluated by comparing its results with the results of simulation for a real-world system as well as systems composed of randomly generated tasks. Experimental results show that our approach can be effectively used in the design evaluation stage of many real-time embedded systems.

Share

COinS