Seminar Title
On Signal Processing Techniques for 5G networks and beyond
Seminar Type
Registration Seminar
Department
Electronics and Communication Engineering
Speaker Name
Anusaya Swain ( Rollno : 520ec1008)
Speaker Type
Student
Venue
EC-303 Seminar Room Department of Electronics and Communication Engineering
Date & Time
08 Aug 2022 05:30 PM
Contact
Prof. Shrishailayya Mallikarjunayya Hiremath
Abstract

Over the past years, the demand to meet the bandwidth requirement has forced us to increase the carrier frequency used for wireless communication. Millimeter/Terahertz band communications are a key enabler for future generation wireless communication systems, allowing high data rates, better physical security, and avoiding electromagnetic waves interference. mmWave/THz and massive multi-input multi-output( MIMO) have improved spectral and energy efficiency. Due to the use of large antenna arrays at the transmitter and receivers, novel signal processing techniques are required to achieve high capacity and reliable communication that cater for mmWave/THz channel conditions. Downlink precoding matrix design depends on the estimates of the downlink channel responses fed back to the base station (BS). Due to a large number of antennas, the channel state information matrix is enormous, leading to excessive feedback overhead. Recent approaches use deep learning (DL) techniques, which compresses the CSI into a codeword with low dimensionality to recover the original channel matrix at the base station. This report proposes a  convolutional neural network-based deep learning technique for compressing the channel response matrix at the user equipment (UE) side. And which is reliably recovered at the base station. New channel models must be developed as the channel characteristics of mmWave/THz frequency bands differ. This report also investigates a THz channel modelling based on machine learning approaches. Channel state information is required for the beamforming mechanism and to design the precoding matrix. Having reliable mmWave/THz channel estimation is challenging due to the presence of a large number of antennas and hence needs to be investigated.