CSCE 990: Real-Time Systems

Projects

The projects in this class are intended to address a number of goals. The primary goal is that they should address a non-trivial aspect of real-time systems. Your project can be either an experimental investigation or a survey or research paper. The project must be a fairly significant piece of work. Projects may be either individual or group efforts. A group project is justified when the scope of the project is unrealistic for a single person within the context of a semester class project.

A secondary goal of the semester project is to synergize with the student's research interests by addressing an aspect of a student's research assistantship or permitting them to do some preliminary investigation of a master's thesis topic, or both. When coordinated with research assitantship work, the project should, ideally, represent an aspect of the RA work which would not have been considered in the context of the project.

A tertiary goal of the semester project is to provide an opportunity to the student to investigate a topic to which they might not otherwise be able to devote time and effort. While ths may be in conflict to some degree with the secondary goal, speculative or fanciful projects often develop into some of the most interesting research projects by discovering or revealing unsuspected potential or relationships among real-time system issues; that's why they call it research.

Project Proposals

You must have your project approved. Thus process is as follows.

1. Email me a one or two paragraph description of your proposed project.
2. If approved, go to step 3. Otherwise, go to step 1.
3. Write a short proposal (about 3-5 pages) that provides more detail. This document will serve as our "contract" for the project.
4. Conduct your research or implement your project.
5. Write a technical report describing your research or project and the results you obtained.

Both the proposal and the final paper should follow the same standard format as a conference or journal paper (about 15 pages, though it may be longer).

Projects Ideas

I encourage you to come up with your own project ideas. However, for those of you panicking because you cannot come up with a project idea, here are some projects you might consider. (Many of the ideas listed below need further explanation, which I will be happy to provide.)

• Solve one of the open theoretical problems highlighted in class.
• Apply real-time scheduling theory analysis to an application executing on a Mote with TinyOS; What task models are appropriate?
• Add new sensors to a Mote and evaluate its temporal requirements.
• Investigate real-time communications protocols for wireless sensor networks.
• Create a temporally-aware anytime algorithm for object tracking using a wireless sensor network.
• Analyze the impact the task model and accompanying scheduling algorithm has on latency in real-time systems specified with a processing graph.
• Enhance a simulator environment that generates random task sets and simulates execution by adding new scheduling algorithms and support for shared resources.
• Implement new sheduling algorithms in Linux, Free BSD, Micro C/OS II, or some other OS and evaluate its performance to support real-time and non-real-time applications.
• Derive equations to bound the memory requirements of real-time systems specified with a processing graph.
• Use Real-Time Java to implement a real-time application.
• Implement a Real-Time application on QNX, Linux, KURT and/or RT-LINUX and compare the performance/predictability of these systems. Good application examples might be multi-media programs, such as VIC or VAT.
• Build a system that supports applications executing in RT-LINUX and KURT at the same time. That is, implement KURT as a non-real-time task in RT-LINUX. Then compare the performance/predictability of an application executing directly on RT-LINUX with the same application executing on KURT (which is on top of RT-LINUX).