Efficient Channel Estimation for Universal Filtered Multicarrier Systems

Abstract

In this paper, we propose an efficient channel estimation (CE) to determine the frequency response of overall subcarriers for universal filtered multicarrier systems, which promises excellent performance without a significant increase in computational complexity. Specifically, based on the even-subcarrier received signal, we design a pilot-assisted scheme combined with a partial inverse fast Fourier transform operation to obtain an estimate of the channel impulse response. Then, to alleviate the power efficiency loss caused by discarding the odd subcarriers, a performance-enhanced CE using the Neumann series expansion technique is proposed. This provides a performance comparable to that of the maximum likelihood-based approach involving a prohibitively burdensome computation. A closed-form mean square error expression and computational complexity analysis are derived to quantify the achievable advantages. Simulation study confirms the effectiveness of the proposed CE.