Research Project

PhonePoint Pen

PhonePoint Pen is a system that uses the in-built accelerometer in mobile phones to recognize human writing. By holding the phone like a pen, user should be able to write short messages or even draw simple diagrams in the air.

PhonePoint Pen: Using Mobile Phones to Write on Air

Sandip Agrawal, Ionut Constandache, Shravan Gaonkar and Romit Roy Choudhury.
In MobiHeld 2009 in conjunction with ACM SIGCOMM 2009, Barcelona, Spain, August 2009.

Phone Point Pen was selected as the winner for Hoffmann + Krippner Award for Excellence in Student Engineering , awarded to the most innovative project in the field of human-machine interfaces (HMI) applications and input devices.

Projects that are Not Really Research

Weight Sensing Alarm Clock [Report]

Designed a Weight Sensing Alarm Clock that effectively gets a person out of bed. The clock used weight sensors under bed-posts to determine if a person was present on the bed. This, in addition to the alarm time was used in deciding if the speaker should beep. When the alarm turned on, it did not turn off until the person was out of the bed. Once the person is out of bed, the clock checks for next 20 minutes if the person returned to the bed. So, the person must stay at least 20 minutes off the bed for the alarm to permanently shut down.

16 Bit Processor

Designed and implemented a complete pipelined processor based on 16 bit MIPS-like architecture on FPGA Board using VHDL. The processor ran the fastest in the Computer Architecture course breaking the modified RC4 algorithm in 7 seconds.

Vending Machine [Report]

Designed a prototype of a vending machine on FPGA board using VHDL that accepts all common coins, has 4 items with distinct prices, returns change intelligently, has a coin return switch to abort a transaction, and displays the total amount the user has inserted. An internal display also keeps track of the number of products available and revenues.

BOE-BOT

Designed a small robot using Basic Stamp. The robot travels a black stripped course counting the number of black and white blocks. It also communicates with other bots to perform the assigned task. Our team won the ECE 2007 Integrated Design Challenge for designing the most functional integrated BOT.