User Tools

Site Tools


finalproject

Final Project

The course final project will be completed individually

Step 1: Literature Survey

For the first step of your project, you will conduct a mini literature survey, selecting three recent research papers on high-level synthesis, and writing a two page report.

Paper Selection

You should choose 3 recent (2010 or later) research papers to read:

  • You can choose to select papers that all focus on a single topic (ie. Scheduling), or different topics.
  • Papers should be “tool” or “architecture” papers, focusing on underlying HLS technologies, not “application” papers that use HLS to accelerate an application.
  • The papers must be at least 4 pages in length.
  • Some possible topics: Scheduling, compiler optimizations, design space exploration, parallel processing, memory banking, debugging, system-level optimization, bitwidth analysis and optimization, resource sharing, power optimization, OpenCL design, etc.
  • Your final project does not need to be related to the papers you select . The purpose of the literature survey is to give you some exposure to current research and spark ideas for the project. However, if you do choose a related area, you may be able to reuse some of your literature survey report in your final project report.

I suggest you find papers by:

  1. Searching on Google Scholar
  2. Checking the program listings for recent FPGA conferences (FPGA, FCCM, FPT, FPL, Reconfig, etc.). The programs will often have one or two sessions dedicated to HLS.
  3. For a given paper, look at what it cites, and what papers cite it. This can also be done using Google Scholar.

Email the instructor the list of your three papers in bibliography format (author names, title, venue name, # pages). The instructor must approve your papers before you proceed.

Survey Report

Write a two page, IEEE double-column format, report discussing the three papers you read. You report should include:

  • The research problems identified by the papers (this may be a common problem if you choose related papers, or may be three different problems if you choose somewhat unrelated papers)
  • The proposed solution, including major assumptions or limitations.
  • Key results presented by the paper.

Survey Presentation

You will give a 3-4 minute presentation (1-3 slides) in class, presenting the problem and solutions for one or more of the papers you read.

Step 2: Proposal

Write a proposal discussing your final project. Proposal requirements:

  • 1 page max
  • Discuss the core idea of your project
  • Discuss how your project fits into current state of the art research, with citations where necessary
  • Include a timeline with weekly milestones

(Unlike the literature survey) You can choose to focus your project on either underlying HLS technologies/tool, or using HLS to accelerate some application.

Part 3: Implementation and Report

In the final, and largest part of the final project, you will carry out your proposed plan and report on the results.

There will be two two deliverable for this part of the project:

  1. A class presentation on either April 15 or 17th. In this presentation, you will describe the problem you are trying to solve, any relevant previous work, and your approach. Present and discuss your results. Like a mini-lecture, your goal is to make sure that the class understands what you are tying to do, and why you are trying to do it. Each presentation will be 8-10 minutes.

2. The final technical report. This should be prepared in IEEE conference format (10pt, two columns), with a 4 page limit. The paper should be of similar writing quality to a IEEE conference paper. Introduce your problem, describe what you implemented, and include results. Use figures where appropriate to help describe your project.

Note: it is not required that you compare the results of your algorithm to those in previous work, however, a very strong project likely would contain at least some comparison to the literature.

Important Dates

  • List of 3 papers for literature survey to instructor: Monday, March 11, 11:59pm
  • Literature survey report: Friday, March 15, 11:59pm
  • Project proposal: Friday, March 22, 11:59pm
  • Final project presentations: In class, April 15 & 17
  • Final project reports: April 17th, 11:59pm
finalproject.txt · Last modified: 2019/02/25 18:08 by jgoeders