Project Policies

Project handouts: Project handouts will be available on the web. The project topic will closely correspond to the lecture topic for that week period.

Working in groups of two: You will work on your projects in groups of two. To forms these groups, I give you two options. First, you can choose grouping on your own. Second, I can choose a partner for you. You may change partners between projects, but not during an individual project. I will probably change partner groups before spring break somepoint, at least once.

The final project could be a larger group. I reserve the right to assign these groups.

If you choose to go alone, I will not make any concessions, but if your partner drops in the middle of a project, then I will take into account

You may discuss the questions in large groups, but each group must independently perform and write-up the required work.

Each Student has a Laptop : Remember GT policy is that everyone has a laptop of a given capability. You will need it throughout this semester.

Experiments: A key part of the Analog IC design, and in particular Neuromorphic Analog VLSI design, experience is the electrical measurements of integrated circuits. We do not want this class to degenerate into a theoretical discussion about circuit analysis. The projects center on experiments to be performed on custom VLSI chips that were built especially for this class.

Further, you will have the unique opportunity to have an analog IC design experience where you will be able to experimentally measure and verify your designs. We require that each group bring in your laptop computer to class. In the past, our approach was using MATLAB as our primary testing engine as well as data analysis, and we had to figure out how to make sure everyone had a working setup with multiple moving parts. Now, we have a new (and improved ) set of tools, based on our experience in MATLAB / Simulink, built on scilab / scicos / Xcos, which is an open source, nearly code compatible, version of MATLAB / Simulink. If you want to generate final data plots in MATLAB, you are welcome to do so.

Therefore, we require that everyone has VirtualBox ( https://www.virtualbox.org/, and you need the extensions file ) as well as have loaded our current version of our Ubuntu virtual machine. You can download the VM (4GB file) from http://users.ece.gatech.edu/phasler/FPAAtool/index.html. Since all of these approaches are available to everyone for free, we expect there will be no hardship for everyone to everything working well. We are using a virtual machine approach to simplify the installation of the resulting system required for testing, characterization, and measurement. If you have interest in running things directly in Ubuntu, we can work with you for the directions for that setup.

Experimental Setup: FPAA devices : This class will make use of experimental results on large-scale Field Programmable Analog Arrays (FPAA), innovated and developed primarily at GT. The entire test setup (for almost all cases) is a laptop computer (which each group will have), working virtual machine setup described above, and an FPAA board. Alot of the design and development can be done before having the board connected, particularly in the design phases, thereby allowing alot of flexibility for each student on completing the resulting assignments. We will be setting up student laptop setups during the first week of classes.

We will start by having FPAA boards in class, and some experimental work might start in class as part of class. In most cases (particularly since the area is empty after class), there is an opportunity to use the boards after class. For the first 3-4 weeks, the experiments are straightforward, assuming one is keeping up with the course material. Once the class starts into design project phase, we will allow boards to be checked out from our class location as long as they are kept in the building. We will have some lectures where we bring additional hardware.

Finally, please convey suggestions about how to improve this laboratory experience for future generations of this class. You are benefiting from the suggestions of earlier years, and I take very seriously any suggestions that make this material more accessible.

There is no need to show any code unless the project explicitly asks for it.

• Deadlines are hard: You need to complete all required items for the project, which will include analysis, experimental measurements, some system level modeling, and data analysis. Projects handed in after this deadline will not be accepted.
• Project Writeup: You will have one writeup, which must be word-processed, two-column IEEE format with data figures integrated into the text. You may not have your data figures added to the end of the writeup. You will submit the writeup as a .pdf file, and only as a .pdf file. You should integrate data and analysis together on your submitted plots. For example, if you perform a curve fit, I expect the curve fit and the data to be on the same graph, where the data points would be in point markers (e.g. "o"), and the curve fit would be a straight line. The plots need to be MATLAB style plots (MATLAB, scilab, python MATLAB), and not other non-technical forms (e.g. Excel). Do not submit your MATLAB code to generate the plot. Your writeup is limited to 9 pages including references and figures. Submit only one writeup per group of two individuals. Make sure both names are on the document.
• Extensions on due dates will not be granted unless there is a very compelling reason (e.g., a medical problem by everyone in the team). To get an extension, you must talk with me before the due date.

Academic Honesty: All conduct in this course will be governed by the Georgia Tech honor code. Additionally, it is expected that students will respect their peers and the instructor such that no one takes unfair advantage of anyone else associated with the course. Any suspected cases of academic dishonesty will be reported to the Dean of Students for further action.