CFO estimation in GFDM systems using extended Kalman filter

Abstract

Multi-carrier modulations are useful in communication systems because of their proper performance in multipath fading channels. The orthogonal frequency division multiplexing (OFDM) is one of the modulations which is used in the most communication systems; however, it does not work properly in many new applications like 5G networks. As a result, several new multi-carrier modulations are presented for 5G systems like the generalized frequency division multiplexing (GFDM). Despite having some advantages to OFDM, the GFDM has some similar problems to the OFDM. For example, the intercarrier interference (ICI) effect which occurs because of existing the carrier frequency offset (CFO). It is caused by dissimilarity of the transmitter and the receiver oscillators and Doppler shift effect. This paper studies the CFO effect on the GFDM performance and applies the extended Kalman filter (EKF) to decrease the CFO and to improve the performance of this system according to the symbol error rate (SER). It is shown that using the EKF can reduce SER more than one order of magnitude which helps GFDM to be practicable for 5G networks.