Nikheel Savant

Aims at predicting the gender of a person based on the language of person’s tweets and images they post in their twitter profile.

Built a Morse Code decoder using polling and interrupts on an input switch.

Synthesized tones of varying amplitudes and frequencies using input capture, output compare, PWM generation, ADCs for sensor input and DACs for driving the actuator.

Objective:
1.Become familiar with Eagle CAD to design the Arduino PCB
2.Assemble the through-hole-equivalent of the Arduino on a breadboard.
3.Install the bootloaders and firmware onto the microprocessors

Modeled/Verified the functionality of a pacemaker and a model heart in UPPAAL. Implemented the pacemaker in mbed‐rtos on LPC1768. Modeled/Implemented a model heart for testing/assurance cases.

Aims at providing a complete control of the Greenhouse over a web application as well as automate the controls using Logistic Regression model in Fully automatic mode.

Developed a mathematical model to calculate the temperature of Brake Fluid without using additional sensors.

Implemented a more efficient sphere decoding using Probabilistic Tree Pruning algorithm and Increasing Radius algorithm on MATLAB.
Analyzed the performance and complexity of sphere decoding. Implemented for 16-QAM and 32-QAM.

Designed an Automatic License Plate Recognition system for real time world.
Implemented using Microsoft Visual Studio and openCV. Verified by recognizing each vehicle license plate whenever it passed through the system.

Designed a system that would recognize what a person is doing. Kinect camera was used as an input to this system.
Implemented using openCV and Microsoft Visual Studio. Verified design by testing it on random sampling of real people.

Designed an Autonomous Parking robot interfacing atmega32 with IR-Sensors and Ultrasonic sensors. The designed robot is capable of finding the empty garages which would be suitable to park. Moreover, Robot would park on its own.

Designed a voice controlled robot interfacing atmega16, MATLAB for voice processing. The designed robot is capable of learning new user voice inputs as well as four directions-Front, Back, Left and Right.
Implemented the design in MATLAB and MATLAB would process voice data and send output to the robot.

- Designed Fetch, Decode, Execute, Memory, Write-back stages of an in-order as well as out of order processor in Java.
- Modeled branch predictors namely, gshare, bi-modal, and tournament predictor as well as modeled split cache design, register renaming, reorder buffer.
- The processor supported perfect memory dependence prediction, branch predictor with a 3-cycle penalty, and cache to handle instruction and code memory access.

Designed an Autonomous Driving Vehicle Prototype in an event organized by Renesas.