Skip to main content

2019 Spring

Semester Short


ATHLETETALK is a mental and behavioral health platform for student athletes. Currently, student athletes attend lectures throughout the semester on these topics. However, attention and retention is low, as these lectures can become rather lengthy. ATHLETETALK strives to fix this issue by bringing the content to a mobile application. Administrators (coaches and other staff) can share articles and videos to the application through the admin website, and athletes can view the information in a more engaging manner (through the Android application).

Simulate, Model, and Emulate for Memory-Centric Architectures

Designing a new server is time-consuming and expensive, and it is not feasible to build a new server every time a new topology is designed. Our projects aims to alleviate that cost by simulating and emulating different server configurations in software, and allowing different software applications traditionally run on the server to run through our emulation. Through our approach, we can see how different components and their latencies affect the performance of different server architectures without the overhead of physically constructing a new server.

Software Bug Triaging

Our project provides a solution to the bug triaging and assignment problem that numerous companies encounter daily. Bugs are often tossed from one developer to another until they can be solved. Our solution aims to eliminate the cost in time and money of tossing by predicting which developer is most likely able to solve a given bug. We utilized multiple different algorithms and test methods, including clustering and natural language processing, in our design to obtain the most accurate results for our sponsor, Dell.

Team Members: 

MLPerf Benchmark Test Suite

Our project involves creating a machine learning test suite that leverages MLPerf (a machine learning benchmarking tool), to be run on Dell servers. The primary objective is to give the user a set of machine learning workloads (which is made up of a machine learning framework and a desired machine learning algorithm) and run the desired workload on the server. Our secondary objective is to output the performance of the server (specifically cpu, gpu, and other power statistics).

Team Members: 

Anthony Chen

Matt Hall

Encrypted Bluetooth Audio Headset

The team created a Bluetooth audio system with high level encryption to allow to users to talk to each other without any fear of the Bluetooth audio being eavesdropped. Currently, Bluetooth devices can be hacked even with their built-in security because of notable flaws when connecting two devices together  Using highly secure 256-bit AES encryption, separate from the Bluetooth system, the team created a signal nearly impossible to decrypt over the air.

DC Motor Boosterpack and Cape

Our team has designed motor controller breakout boards for the BeagleBoard PocketBeagle and the Texas Instruments MSP432 LaunchPad. The goal of our project is to allow users of these boards to easily interface with and drive DC brushed, stepper, and servo motors. Each of our motor driver breakout boards has the appropriate hardware peripherals to drive the motors, is stackable, and is divided into four modules that allow for parallel motor control.

Wafermap Defect Pattern Recognition

The manufacturing yield of semiconductor under silicon technology depends heavily on wafer processing. The wafer map may not only reveal a specific defect pattern for engineering trouble shooting but also provide an opportunity for process control, test optimization, quality improvement.  However, reviewing all the wafer maps could be a tedious task, if done manually. This project is to derive an automated and efficient way to classify the wafer map into signature patterns.

Texas Instruments IoT Development Board

The field of the “Internet of Things” (IoT) consists of any items with wireless communication capabilities. This includes devices like smart TVs, smartphones, some wireless home security systems, and much more. Many of these devices communicate across frequencies like Bluetooth and WiFi. Because these networks are having to support an exponentially increasing number of smart devices, their frequencies are becoming cluttered and the risk of becoming overloaded and slowing down continues to increase.

Encouraging Sustainable Mobility through Games in the Elementary Classroom

In the digital age of technology, children are becoming less mobile and active in their daily lives. Our project aims to use technology to encourage, rather than hinder, healthy habits in elementary-aged students through an interactive, visual, and educationally supportive web application. Through close collaboration with current elementary teachers, students, and industry experts, our team has developed an efficient system consisting of an easy-to-use RFID badging interface for students and an engaging game design which supports the classroom curriculum.

C2000 Based TI-RSLK

C2000 platform is a popular micro-controller family for power electronics and control applications. It is also one of the few TI platforms that support Matlab. The goal of this project was to take the TI robotics system learning kit (RSLK) and switch from using the MSP432 to using the C2000. The software for this project has been written in Matlab, and it utilizes all the sensors — motors, bumper switches, IR sensors, and line sensor — in the basic RSLK in order to successfully navigate through a maze.

Team Members: 

Ali Kedwaii

Subscribe to 2019 Spring