A Generalized Two-Dimensional FFT Precoded Filter Bank Scheme With Low Complexity Equalizers in Time-Frequency Domain

Recently, the two-dimensional (2D) fast Fourier transform (2D-FFT) filter bank (FB) technique was proposed as an alternative to new wireless communication technologies covering several application scenarios. This technique presents similar characteristics to orthogonal time-frequency space modulation (OTFS), such as lower PAPR and robustness in high mobility scenarios; in addition, thanks to the use of filter bank modulation, the technique has a good spectral localization. In this work, we present the generalization of the 2D-FFT FB system by making changes in the waveform structure to obtain advantages in the symbol detection process. The proposed idea allows the use of low-complexity equalizers in the time-frequency (TF) domain, bringing a good performance/complexity trade-off and obtaining bit error rate (BER) results of the same order than delay-Doppler domain receivers for OTFS. Theoretical and simulation results show that it is possible to obtain a good error performance in high mobility scenarios even using simple equalizers in the TF domain combined with waveform structure adjustments.