Multiuser Interference Cancellation for GFDM With Timing and Frequency Offsets

Abstract

In uplink multiuser systems, different timing offset (TO) and/or carrier frequency offset (CFO) of multiple users cause multiuser interference (MUI) to a desired user. In this paper, we propose two different MUI cancellation methods, namely weighted parallel interference cancellation (WPIC) and adaptive interference cancellation filter (AICF), for uplink transmission with generalized frequency division multiplexing (GFDM) waveform over Rayleigh fading channel given the existence of TO and CFO of multiple users. The proposed WPIC performs MUI estimation using the detected data symbol of other users and obtains the optimal weight to maximize the signal-to-interference ratio (SIR) of the desired user. We also propose the AICF that aims to maximize the SIR of the desired user. Based on our SIR expression derived in closed-form in the presence of TO and CFO, the AICF is obtained by solving the Rayleigh quotient problem. Lastly, the impact of multiple antennas at the receiver on MUI cancellation is also analyzed to improve reliability. The simulation results show that the proposed WPIC and AICF effectively eliminate MUI and that the AICF can outperform the WPIC in terms of the SIR and spectral efficiency. In addition, the multiple receiver antennas effectively cancel MUI as the number of antennas goes to infinity.