Turbo Receiver Channel Estimation for GFDM-Based Cognitive Radio Networks

Abstract

Generalized frequency division multiplexing (GFDM) is a promising candidate for 5G waveforms because of its flexibility to meet the requirements of different scenarios and applications. Compared with the orthogonal frequency division multiplexing, the lower out-of-band radiation and higher spectrum efficiency make GFDM the qualified waveform solution for cognitive radio (CR). In CR networks, channel estimation is the necessity and premise to guarantee reliable spectrum sensing. Because of the block modulation structure for GFDM, conventional channel estimation algorithm cannot be applied to GFDM directly. Also, the feedback information in channel decoding is not fully utilized by the conventional algorithm. To solve these problems, in this paper, a modified turbo receiver is designed to utilize the feedback information for channel estimation. The modulator for the pilot insertion is also improved to make pilots orthogonal to data subcarriers. Based on the modified turbo receiver, an iterative channel estimation strategy combined with threshold control is proposed to cope with the noise enhancement. The symbols rebuilt for the iterative channel estimation are verified that only the credible symbols can be retained for the next iteration. Simulation results demonstrate that the proposed channel estimation method has the better bit error rate and mean-squared error performances over Rayleigh fading channel compared with the conventional algorithm.