Hi! I’m Sima Mitra and I am a recent Electrical and Computer Engineering graduate. I received my Masters of Engineering in May of 2013 and my Bachelors of Science in 2012 both from Cornell University in Electrical and Computer Engineering. I’m an interdisciplinary electrical engineering with interests in algorithm development, embedded design, photonics, robotics, and 3D printing.
- Hardened the embedded C firmware to create a robust data uplink from sensors to central hub to server.
- Increased wireless pressure sensor battery life 6x by implementing successive back-off when sensor was unresponsive.
- Circuit design, layout, and assembly of rechargeable battery backup for WiFi product..
- PCB Layout for new -40°C tolerant BLE cold-chain monitoring product.
- Board design and firmware development for new Oxygen sensor.
- Designed an extensive automated firmware-testing platform in C#. The platform automatically programmed devices, interacted with debug ports, controlled sensors, and read LED blink patterns.
- Successful FCC/IC Certification of an OEM ZigBee module.
- Created a python tool that automatically diagnosed the cause of missing or late sensor readings by parsing device packets.
- Developed pulse transit time algorithms in MATLAB to estimate blood pressure.
- Analyzed datasets to understand factors that affected biological signal morphology and quality.
- Evaluated electrocardiogram (ECG) and electrodermal activity (EDA) sensors on the wrist.
- Evaluated potential partner companies’ photoplethysmography (PPG) sensors.
- Collaborated with the product development team to preserve signal quality during many prototype iterations.
- Defined test protocols and developed test rigs to evaluate PPG sensor performance.
- Researched optical properties of human skin to develop suitable human analogs for sensor testing.
- Wrote computer vision algorithms to automatically assess light output shape and intensity of different sensing elements.
M.Eng. in Electrical and Computer Engineering
Graduated May 2013
B.S. in Electrical and Computer Engineering
Graduated cum laude in May 2012
TicTocTrac / Quantified-Self wristwatch that measures your perception of time.
Designed an LED wristwatch based on the Atmega 32U4 microcontroller. All software written in C and is open source. TicTocTrac uses a Real Time Clock, microUSB, microSD, a rechargeable Lithium battery, and vibration motor all on its own custom designed PCBs. Featured on the Forbes, Hackaday, Makerbot, Adafruit, and Arduino Blogs. TicTocTrac is now an Open Source Hardware project that anyone can build, modify or improve. More at www.tictoctrac.com.
Autonomous Quadcopter Docking System / System Integration
For my Masters of Engineering project, I created a system that would allow a quadcopter to autonomous locate and land on a ground station. The purpose of this system was to outline the framework for a quadcopter based data collection or surveillance system that copes with the relatively short battery life of these highly mobile devices by consistently landing the UAV safely in a designated location to be recharged. The 3D Robotics ArduCopter was chosen as the quadcopter platform since it is durable, capable of autonomously hovering in place, and is capable of carrying a payload, such as the camera used to determine the location of the dock. The system was then devised such that the quadcopter can correctly determine the location of a target ground station while hovering and then land when on the target. Only commercially available components and free software was used so that the entire docking system is easily accessible to future researchers and UAV enthusiasts.
Nintendo Ninja / A Hardware-Based FPGA AI for Super Mario Bros.
Helped design a hardware AI on the Altera Cyclone II FPGA that could play World 1-1 of Super Mario Bros. by emulating a controller. More at www.nintendoninja.com.
Autonomous Mobile Robotics Competition / Algorithm design in MATLAB
Designed a SLAM algorithm for an iRobot Create to navigate an unknow area and search for markers.
Fab@Home 3D Printing student project team/ Electronics team; Host and Embedded Software Design
Worked on interfacing between the controlling computer and the motherboard of the Model 3 Fab@Home 3D printer.
Automated segmentation of adult Tachycineta bicolor from images in their nests. / Computer Vision detection of tree swallows
Designed a computer vision algorithm that located and segmented the outline of an adult tree-swallow from videos taken of their nests. This final project won Best Poster and Best Poster Presentation in the class.
1.25 Tb/s nanophotonic interconnection for high bandwidth, short range computing / Nano Optics
Researched and designed a 1.25 Tb/s nanophotonic interconnection for high bandwidth, short range computing.
Effects of Waveguide Shape on Bending Loss / Fiber and Integrated Optic
Researched, designed and tested in a lab a novel silicon photonics waveguide structure with reduced bending losses.
Stable Ring Laser / Lasers and Optoelectronics
Designed a stable ring laser in Mathematica and testing it in a lab.
Software I use
- OrCAD Capture
- OrCAD PCB
I also have experince with Eagle, Eagle3D, COMSOL, L-Edit, LabVIEW, and Java.
Skills & Interests
- Algorithm Development
- Autonomous Robotics
- Computer Vision
- Circuit Design
- PCB Layout
- Embedded Systems
- Photonics and Fiber Optics
WAn FPGA-based AI that uses video input from an NES console to automatically play the game Super Mario Bros. All of the video analysis and AI techniques are performed using Verilog-compiled hardware running on an Altera DE2 Cyclone board. The project combines NTSC decoding, VGA output, kernel-based pattern matching, real-time image manipulation, and NES controller emulation.
This project was created as a final project for ECE5760 at Cornell University in Spring 2013.
Dont' just tract time, track your perception of time!
TicTocTrac is a watch designed as part of ECE4760 at Cornell University that allows you to not only keep time, but track you perception of it. TicTocTrac is an open source project that anyone can build and improve on!
Quadcopter Landing Under Camera Control
The goal of this project was to design the systems and algorithms necessary to allow a quadcopter to autonomous locate and land on a station target. The purpose of this system was to outline the framework for a quadcopter based data collection or surveillance system that copes with the relatively short battery life of these highly mobile devices by consistently landing the AAV safely in a designated location to be recharged. A system was devised such that the quadcopter can correctly determine the location of a target ground station while hovering and then land when above the target. Only commercially available components and free software were used to so that the entire docking system is easily accessible to future researchers and UAV enthusiasts.
Autonomous iRobot Create Navigating Final Competition Arena
An autonomous iRobot Create localizes itself using a particle filter and sonar data while searching the competition areana for black squares on the ground.
On-chip photonic system capable of 1.25 Tb/s data transfer, designed using pervious research. This system would be which is an improvement over previously demonstrated 1 Tb/s systems.
Created by: Sima Mitra, Matthew Storey, Jason Wright, and Logan G. Wright. Image credit: Matthew Storey
A NIOS II-based method for computing the Mandelbrot set at an arbitrary zoom level. The Mandelbrot set was then displayed at 640x480 resolution using VGA in 8-bit color where the color of a region corresponds to the number of iterations computed before the set diverged.Regions inside of the Mandelbrot set were white, while outside of the set the colors ranged from blue to red for 1 to a maximum of 256 step convergence.
Created a computer algorithm that used local binary patterns (LBP) to analyze videos of the nests of Tachycineta bicolor (tree swallows) to determine when the adult tree swallows were present in the nest. These videos were record by the Cornell Lab of Ornithology for an on-going study on climate change an it’s effect on these birds. This algorithm could be used to drastically reduce the man-hours needed to review hundreds of hours of these videos.
Created by Sima Mitra and Yichi Zhang