Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/612438
Title: Estimation and Synchronization for Single Carrier and OFDM Systems Using Neural Network
Researcher: Chaudhari, Mahesh Shamrao
Guide(s): Majhi, Sudhan
Keywords: Engineering
Engineering and Technology
Engineering Electrical and Electronic
University: Indian Institute of Technology Patna
Completed Date: 2023
Abstract: The adaptive communication system is going to play a significant role in fifth-generation newline(5G) and beyond wireless communication, where the physical layer signal parameters need newlineto be changed at the transmitter per system requirement, and the receiver needs to estimate newlinethem to recover the signal. The intermediate frequency (IF) carrier, symbol rate, symbol newlinetiming offset (STO), and carrier frequency offset (CFO) are the most crucial parameters in newlinean adaptive communication system, whether it is a single carrier or multi-carrier system, for newlineautomated and reliable detection. The symbol rate must be known to the receiver prior to newlinesymbol timing recovery, known as STO estimation. In addition to this, the desired IF carrier newlineinformation is also needed to convert the passband signal to an error-free baseband signal. newlineIn multi-carrier systems such as orthogonal frequency division multiplexing (OFDM), mul- newlinetipath delay introduces the STO in the system, due to which inter-symbol interference (ISI) newlinearises in the system. Also, due to Doppler shift because of motion between transmitter and newlinereceiver or frequency mismatch between the oscillators of transmitter and receiver, CFO newlinegets induced in the OFDM system, which destroys the orthogonality among the subcarriers. newlineSo, these parameters need to be estimated and compensated for decoding the transmitted newlinemessage. newlineThe above estimation problems can be solved using artificial intelligence (AI)/deep newlinelearning (DL) approaches and statistical approaches. In the first research problem, we have newlineestimated IF carrier using a deep neural network (DNN). In statistical-based IF carrier esti- newlinemation methods, signal bandwidth must be known prior to the IF carrier estimation, but the newlineproposed scheme does not require knowledge of it. The proposed scheme also does not re- newlinequire any information regarding modulation schemes, channel state information (CSI), and newlinesynchronization parameters. The performance of the proposed scheme is evaluated in the newlinepresence of Rayleigh and Rician fading environments for five
Pagination: xxiii, 135p.
URI: http://hdl.handle.net/10603/612438
Appears in Departments:Department of Electrical Engineering

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File178.66 kBAdobe PDFView/Open
02_prelim pages.pdf95.5 kBAdobe PDFView/Open
03_content.pdf24.98 kBAdobe PDFView/Open
04_abstract.pdf20.57 kBAdobe PDFView/Open
05_chapter 1.pdf359.16 kBAdobe PDFView/Open
06_chapter 2.pdf228.16 kBAdobe PDFView/Open
07_chapter 3.pdf368.05 kBAdobe PDFView/Open
08_chapter 4.pdf739.72 kBAdobe PDFView/Open
09_chapter 5.pdf288.56 kBAdobe PDFView/Open
10_chapter 6.pdf413.24 kBAdobe PDFView/Open
11_chapter 7.pdf595.12 kBAdobe PDFView/Open
12_chapter 8.pdf27.56 kBAdobe PDFView/Open
13_annexures.pdf55.09 kBAdobe PDFView/Open
80_recommendation.pdf259.85 kBAdobe PDFView/Open
Show full item record


Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial 4.0 International (CC BY-NC 4.0).

Altmetric Badge: