Wavelet-Aided OTFS With Improper Gaussian Noise

Abstract

Sixth-generation (6G) wireless communication is expected to include much higher mobility usage scenarios compared with previous generations. This situation brings with it challenges such as Doppler and hardware-induced distortion effects, which traditional methods are insufficient to handle. In this study, wavelet-aided orthogonal time-frequency space (W-OTFS) modulation, known for its resistance to Doppler shifts, is addressed under the effect of improper Gaussian noise (IGN) originating from hardware faults, which is largely neglected in the literature. Considering a highly mobile 6G scenario, a hard-decision-based widely linear minimum mean square error (WL-MMSE) detector is designed to evaluate the bit error rate (BER) of the proposed W-OTFS system, and the simulation results are presented. In addition to Daubechies and Haar wavelets, which are frequently preferred in W-OTFS applications, results obtained with different orthogonal wavelet types are also introduced. The extensive simulation results demonstrate that the W-OTFS provides more than 5 dB gain for achieving a BER value of $10^{-2}$ compared with the traditional OTFS, regardless of the wavelet type. Furthermore, W-OTFS is proven to resolve the high PAPR issue in OTFS. Depending on the selected wavelet type, a minimum PAPR reduction of 2.5 dB was achieved compared with OTFS.