Deep Learning-Enabled Signal Detection for MIMO-OTFS-Based 6G and Future Wireless Networks

TL;DR

The paper tackles the high computational burden of signal detection in MIMO-OTFS systems operating in high-m Doppler channels by proposing low-complexity deep neural network detectors. It compares MLP, CNN, and ResNet architectures against a traditional MRC-assisted maximum likelihood detector under Nakagami-$m$ fading, showing that the MLP offers near-MLD BER with linear $ ext{O}(MN)$ inference complexity. The work provides detailed complexity analyses and extensive BER simulations in SISO and 2×2 MIMO scenarios, demonstrating that DL detectors can achieve comparable performance with far lower online complexity, particularly in high-throughput 6G-like settings. It also discusses practical implications, noting that preprocessing still dominates overall receiver complexity and outlining future directions for end-to-end learning, RIS-OTFS, ISAC-OTFS, and distributed/advanced training paradigms.

Abstract

Orthogonal time frequency space (OTFS) modulation stands out as a promising waveform for sixth generation (6G) and beyond wireless communication systems, offering superior performance over conventional methods, particularly in high-mobility scenarios and dispersive channel conditions. Recent research has demonstrated that the reduced computational complexity of deep learning (DL)-based signal detection (SD) methods constitutes a compelling alternative to conventional techniques. In this study, low-complexity DL-based SD methods are proposed for a multiple-input multiple-output (MIMO)-OTFS system and examined under Nakagami-$m$ channel conditions. The symbols obtained from the receiver antennas are combined using maximum ratio combining (MRC) and detected with the help of a DL-based detector implemented with multi-layer perceptron (MLP), convolutional neural network (CNN), and residual network (ResNet). Complexity analysis reveals that the MLP architecture offers significantly lower computational complexity compared to CNN, ResNet, and classical methods such as maximum likelihood detection (MLD). Furthermore, numerical analyses have shown that the proposed DL-based detectors, despite their low complexity, achieve comparable bit error rate (BER) performance to that of a high-performance MLD under various system conditions.

Figures (8)

• Figure 1: MIMO-OTFS system model.
• Figure 2: MLP architecture.
• Figure 3: CNN architecture.
• Figure 4: ResNet architecture.
• Figure 5: Computational complexity comparison for massive MIMO configurations ($M = 128$, $N = 128$) with 256-QAM and 1024-QAM modulation. MLD complexity scales with $N_T$ while DL-based detectors maintain constant complexity.