FBMC Modulation Signal Detection Based on Gated Recurrent Unit

Abstract

Filter bank multicarrier (FBMC) modulation is a promising candidate modulation method in future communication systems. Aiming at the problems of imaginary part interference, channel estimation and symbol detection in multi-carrier (FBMC) systems, a detection method for FBMC systems based on gated cyclic units is proposed. A simulation system based on the combination of the GRU(Gated Recurrent Unit) model and the offset quadrature amplitude modulation (OQAM)-FBMC system is completely built to optimize the preprocessing of the FBMC demodulation data. The NN-GRU joint signal detection system and the CNN-GRU joint signal detection system are used for integrated modeling of channel estimation, symbol detection, OQAM demodulation and other modules. This method not only improves the structure of the traditional FBMC model, but also optimizes the reception of the traditional model. The parameters of the module. Compared with traditional classifiers, the accuracy of symbol detection is improved. Theoretical analysis and simulation results show that the bit error rate (BER) performance of the new method is significantly improved compared to the performance of the pilot estimated FBMC system.