Research

I have taken part in several different projects during my time at the University of Maryland and also at IITM. Here is a brief summary of the research conducted in these projects, starting from the oldest.

Indian Institute of Technology, Madras

  • Bachelor’s Thesis Research

    • Analyzed the stability, rate of convergence and performance of communication systems using control theoretical concepts, by modeling the system by linear delay differential equations.
    • Investigated congestion control in Internet from a control theoretical perspective: understanding the impact of buffer size and Active Queue Management algorithms on stability and congestion in the network.

University of Maryland, College Park

  • Delay Tolerant Networks

    • Investigated Delay Tolerant Networks (DTNs), to develop analytical models to understand the Epidemic Routing protocol.
      • Evaluated several mathematical models using discrete time equivalents of continuous time Markov processes as the basis, such as jump Markov processes. Simulated the efficiency of the models using MATLAB.
  • Master’s Thesis Research

    • Surveyed established methods of blind equalization in wireless communication systems to develop a novel “Blind receiver” for LTE systems.
      •  Identified several blind equalization techniques for application in LTE.
      • Designed MCMA/AMA equalizer algorithm for a 2×2 MIMO-OFDM system in MATLAB. Compared performance with traditional non-blind equalizers.
    • Developed novel techniques to estimate Precoder matrix used in LTE transmission
      • Implemented three different methods to estimate, at the receiver, the precoder matrix used by an LTE transmitter, namely – Blind equalization using MCMA/AMA, Hypothesis testing and using Cluster Variance.
      • Observed a favorable trade-off between error performance and computation complexity with a simplified hypothesis testing method, compared to other techniques.
  • Real-Time Video Rate Control Research

    • Explored the design of video-quality aware rate control algorithms for real-time video conferencing applications, to optimize end-user experience, and reduce network congestion.
      • Designed a video rate adaptation policy using a Markov Decision Process (MDP).
      • Proposed to estimate the optimal policy using Q-learning.