On Channel Estimation in Universal Filtered Multi-Carrier (UFMC) System

Abstract

Universal Filtered Multi-Carrier (UFMC) is a key technique of multi-carrier transmission to meet high data rate requirement of wireless communications. As of some limitations in orthogonal frequency division multiplexing (OFDM), it is not appropriate for 5G. Being an entrant of 5G, UFMC merge the simplicity of OFDM with robustness of filter bank multi-carrier (FBMC). Channel estimation is one of the prominent research topics of UFMC modulation technique. Channel state information (CSI) derived from channel estimation is essential for receivers to perfectly recover the transmitted signals. There are many well defined classical pilot-aided channel estimation techniques for OFDM, Due to the similarities of UFMC with OFDM; it is also called UF-OFDM. Therefore with some modifications, the channel estimation techniques used for OFDM are applicable to this 5G technique. Least Square (LS), Minimum Mean Square Error (MMSE) are two different channel estimation techniques which can be evaluated for UFMC. Interpolation techniques such as Linear and Spline are also evaluated for UFMC. Simulation results show that MMSE provides better MSE performance than LS, since at higher SNRs LS is more vulnerable to noise. The MSE floor is observed in AWGN and Rayleigh fading channels.