Basic Information
- Project Title / Domain: Estimation of MIMO OFDM wide-band channels using Matrix All-Pass IIR Filters
- Name: Rishabh Shetty
- Guide: Kumar Appaiah
Description of Abstract / Project
Limited-feedback Channel State Information (CSI) has made channel estimation a necessity in modern wireless communication systems, especially after the advent of 5GmmWave networks, which operate across wide-band frequencies to achieve high data-rates and low latency. For achieving the best possible data-rate it requires one to reconstruct a channel response from the pilot symbols across allocated bandwidth which most accurately matches the true channel behavior. A Precoder for transmit vector derived from the true channel response sets an upper limit for the achievable rate at every subcarrier frequency. I worked on reconstruction of unitary precoder matrices across entire frequency bandwidth using Unimodular Interpolation and the need of choosing a unique set of right singular vectors from Singular Value Decomposition (SVD) of channel response matrix for closest approximation to true channel response. A matrix All-Pass filter is designed which fits the continuous frequency response of the entire wide-band channel from which the precoders at required frequencies get sampled. Such an implementation can be realised as a linear constant coefficient difference equation (LCCDE) at transmitter and receiver.
Whom did you work with?
The prof directly
Relevant Courses
EE229, EE341, EE338, EE764
How did you connect with the prof?
I mailed him mentioning the work I have done in the domain and then he scheduled a meet to discuss more, we decided the SRE project after some initial meets.
Expectations from Prof
He expected me to have good grip over fundamentals of CSP. He also asked me to study the course contents covered in a first course on wireless communication which includes concepts like Channel fading, OFDM, MIMO systems and some idea about 5G specs was useful for keeping in mind the practical implications of the project. I expected to work on something which would be useful for modern wireless communication. In my internship I worked on a 5G modem chip, so working on a project in the same domain would help me ahead.
Logistics of the Project
I gave a presentation at the end of semester and submitted my report for the project. I used to meet him once or twice in a week in an hour long session to discuss the progress and the work ahead.
Reading Material
Mostly read papers related to the previous work done by him on the project, and read Tse and Vishwanath Wireless Communication for understanding some concepts related to modern communication.
Tools / Simulations / Softwares / Hardwares Used
I did my simulations on matlab/python, mostly the work was theoretical
Outcomes / Deliverables
It was a head start for the DDP work and also inspired an idea for writing in ISP letters
Extension to DDP
Yes I did, I am continuing the work in my DDP as its related to my domain and the prof is really enthusiastic about the project and gives a lot of time for discussions
Type of Project
theoretical
Connection to DDP
nan
Advice
You can take projects under him if you are interested in wireless communication, he is very enthusiastic about the work and will give you a lot of time