CV

M.S.

Stanford University

Dept. of Electrical Engineering

B.Tech

Indian Institute of Technology Dharwad

Dept. of Electrical, Electronics, and Communication Engineering

• Digital Systems and Lab, Digital Signal Processing, Microprocessors and Microcontrollers, Digital Communication and Coding Theory, Wireless Communications, VLSI Design, Embedded Systems Design and Lab, Next-Generation Wireless Networks, Mathematics for Data Science, Probability Models and Applications, Pattern Recognition and Machine Learning, Optimization Theory and Algorithms.

President of India Gold Medal

IIT Dharwad

Awarded to the student demonstrating outstanding academic performance.

Age of Information Minimization in Goal-Oriented Communication with Processing and Cost of Actuation Error Constraints

arXiv (submitted to IEEE Transactions on Communications; under-review)

We study a goal-oriented communication system in which a source monitors an environment that evolves as a discrete-time, two-state Markov chain. At each time slot, a controller decides whether to sample the environment and if so whether to transmit a raw or processed sample, to the controller. Processing improves transmission reliability over an unreliable wireless channel, but incurs an additional cost. The objective is to minimize the long-term average age of information (AoI), subject to constraints on the costs incurred at the source and the cost of actuation error (CAE), a semantic metric that assigns different penalties to different actuation errors.

Karush-Kuhn-Tucker Condition-Trained Neural Networks (KKT Nets)

arXiv

This paper introduces a new approach to convex optimization problems, based on the fact that any set of primal or dual variables satisfying the Karush-Kuhn-Tucker (KKT) conditions is necessary for optimality. The approach uses Theory-Trained Neural Networks (TTNNs) and a KKT Loss loss function to measure the network's outputs. The study shows that minimizing the KKT Loss alone outperforms training with a weighted sum of KKT Loss and Data Loss.

Learning Short Codes for Fading Channels with No or Receiver-Only Channel State Information

arXiv

This study designs short-length codewords for next-generation wireless networks that don't use channel state information (CSI) or rely solely on CSI at the receiver (CSIR). The autoencoder architecture is used to achieve mutual orthogonality in the no-CSI case, but not in the CSIR-only case. The codes perform better than classical codes in both no-CSI and CSIR-only cases.

DVB-S2X Implementation

Digital Video Broadcasting using Satellites Version 2 extension. ETSI standards implementation on a FPGA board. A part of ISRO RESPOND project. Supervised by Prof. Rajshekhar V. Bhat, IIT Dharwad

• Jointly developed a primitive FPGA implementation in Verilog of the transmitter sub-system. Responsible for CRC-8 Encoder, Scrambler, BCH Encoder and other minor modules.
• Skills: Digital System Design, FPGA, HDL Coding, Matlab Coding and scripting
• Tools and Hardware: Vivado by AMD, Pynq-Z2 Development Board

Ultrasonic Sensor-based Parking Assist System

EE 615 Embedded Systems Design Course Project, IIT Dharwad

• Developed a automotive parking sensor using TI TM4C123GH6PM microcontroller. The distance was captured using Ultrasonic sensor and displayed on small monochrome OLED display. Drivers for the OLED display were written from scratch.
• Embedded C coding (for Tiva Board), I2C communication for SSD1306-based OLED.
• Tools and Hardware: Code Composer Studio, Tiva C series TM4C123G LaunchPad Board, HC-SR04 sensor, 0.96in SSD1306 OLED display.