Estimation and Synchronization for Single Carrier and OFDM Systems Using Neural Network

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