PaperlayerView exact revision on arXiv

Source-linked AI summary

Comparative Performance Analysis of OTFS and OFDM Modulations for Mobile Wireless Communications

J. Marcos Leal B. Filho, Danilo de S. Pena, Marília C. Muniz, Álvaro A. M. de Medeiros, Vicente A. de Sousa Jr

arXiv:2609.11623v1cs.NI

TL;DR

The paper addresses the need for clearer quantitative boundaries on when OTFS or OFDM performs better in mobile wireless scenarios. It compares the schemes across mobility, modulation order, multipath, and equalization, finding OTFS advantages in high-mobility and multipath settings, with the gap depending on modulation order.

  • Problem

    Quantitative boundaries for when OTFS or OFDM is superior remain underexplored, despite OFDM's susceptibility to Doppler-induced interference in high-mobility channels.

    Source: §I · pp. 1–2

  • Method

    The paper quantitatively compares OTFS and OFDM across modulation orders, mobility levels, multipath conditions, and equalization techniques under perfect CSI.

    Source: §I · p. 2, §V · p. 5

  • Results

    OTFS outperforms OFDM in high-mobility and multipath scenarios, especially at 4- and 16-QAM; at 64- and 128-QAM, its advantage is limited to high SNR, while Message Passing gains over LMMSE from 0–21 dB.

    Source: §V · p. 5

  • Takeaways & Limitations

    The evaluations delineate where each modulation scheme and configuration excels, with OTFS showing stronger Doppler robustness in the reported high-mobility settings.

    Source: §V · p. 5

Abstract

from arXiv · show

This paper provides a quantitative performance comparison between Orthogonal Time Frequency Space (OTFS) and Orthogonal Frequency-Division Multiplexing (OFDM) modulation schemes, focusing on mobile wireless communication scenarios. We evaluate and compare both schemes based on critical communication scenarios and configurations such as mobility levels, modulation orders, multipath environments, and equalizers. The study systematically identifies conditions where OTFS and OFDM each exhibit optimal performance. Results from simulations demonstrate that OTFS outperforms OFDM consistently for high mobility scenarios and multipath environments. Depending on the modulation order, the performance gap between OTFS and OFDM might be very close or many orders of magnitude. Moreover, at low and mid values of SNR, the non-linear equalizer performs better than traditional linear equalizers for OTFS.

I. INTRODUCTION

The introduction frames high mobility as a challenge for OFDM and presents a quantitative OTFS–OFDM comparison to clarify which configurations perform better. The study varies modulation order, mobility, multipath, and OTFS equalization under perfect CSI.

  • High Doppler shifts cause inter-carrier interference that degrades OFDM in rapidly time-varying, high-mobility channels.Source: §I · p. 1
  • OTFS maps symbols to a delay-Doppler grid, providing time-frequency diversity and robustness to delay and Doppler spread.Source: §I · p. 1
  • Prior work examines OTFS detection and BER across Doppler, multipath, realistic RF impairments, diversity, and high-speed settings.Source: §I · pp. 1–2
  • The paper addresses underexplored quantitative boundaries of scheme superiority by comparing OTFS and OFDM under perfect CSI across 4-QAM to 128-QAM, 0 to 380 km/h, 2 to 8 paths, and LMMSE versus Message Passing equalization.Source: §I · p. 2

II. OFDM AND OTFS

This section introduces the basic theory of OFDM and OTFS and their Fourier-based transformations and modulation parameters.

  • The section covers OFDM and OTFS fundamentals, including Fourier-based transformations and modulation parameters.Source: §II · p. 2

A. Orthogonal Frequency Division Multiplexing (OFDM)

OFDM represents information in a time-frequency matrix and transforms it into a transmitted time-domain signal. The section also describes its high-Doppler vulnerability and received-signal equalization.

B. Orthogonal Time Frequency Space (OTFS)

OTFS represents the channel in the delay-Doppler domain, where time-varying multipath channels become nearly static. Its modulation and equalization choices trade implementation cost and detection performance.

B. Orthogonal Time Frequency Space (OTFS)

  • OTFS transforms the channel representation into the delay-Doppler domain, making time-varying multipath channels nearly static.The paper describes this representation as supporting robustness and reliability in rapidly varying channels.Source: B. Orthogonal Time Frequency Space… · p. 2
  • The ISFFT route maps delay-Doppler symbols to the time-frequency plane before Heisenberg transmission, while IDZT offers an alternative intended to reduce computational cost.At reception, Wigner and SFFT transforms return the signal to the delay-Doppler domain.Source: B. Orthogonal Time Frequency Space… · p. 2
  • For OTFS, LMMSE regularizes inversion to mitigate ISI and limit noise amplification, while large sparse channel matrices can motivate iterative MP solutions.The stated reason for using iterative, nonlinear approaches in practical scenarios is to avoid direct matrix inversion.Source: B. Orthogonal Time Frequency Space… · p. 3
  • MP iteratively updates symbol posterior probabilities through factor-graph messages and can improve detection over LMMSE in highly dispersive channels.The gain comes with higher computational complexity, and convergence may depend on channel sparsity and iteration count.Source: B. Orthogonal Time Frequency Space… · p. 3

III. SYSTEM MODEL

The system model compares OTFS and OFDM using matched parameters and a synthetic tapped-delay-line channel. The configuration fixes receiver CSI assumptions and varies modulation order.

III. SYSTEM MODEL

  • The simulations assume perfect receiver CSI, use zero-padding guard intervals, and set the numbers of subcarriers and OTFS symbols to 32.The channel is synthetically generated in the delay-Doppler domain using a tapped-delay-line model with random path delays, Dopplers, and complex gains.Source: §III · p. 3, Table I · p. 3
  • OTFS and OFDM use identical underlying parameters, including a 4 GHz carrier frequency and 15 kHz subcarrier spacing, for a fair comparison.The OTFS system uses the Zak transform scheme.Source: §III · p. 3
  • The modulation-order comparison covers 4-QAM, 16-QAM, 64-QAM, and 128-QAM to assess performance under high Doppler conditions.The system model section frames the evaluation around simulation configuration, scenarios, system parameters, and performance metrics.Source: §III · p. 3

A. Evaluation Scenarios

The evaluation varies user speed across static, intermediate, and high-mobility scenarios while retaining the other physical channel-model characteristics.

A. Evaluation Scenarios

  • At 0 km/h, the static channel is nearly time-invariant over the 2.13 ms frame, with negligible Doppler effects.Source: A. Evaluation Scenarios · p. 4
  • At 127 km/h, intermediate mobility creates moderate Doppler spread, introducing selectivity and inter-carrier interference in OFDM.Source: A. Evaluation Scenarios · p. 4
  • At 380 km/h, high mobility produces severe Doppler effects and fast channel variations within a single frame.Source: A. Evaluation Scenarios · p. 4

B. Performance Metrics

The evaluation uses bit error rate across SNR values to compare robustness and investigate where performance changes between modulation schemes.

B. Performance Metrics

IV. RESULTS

The results compare OTFS and OFDM BER across modulation orders, mobility, multipath, and equalization. For modulation orders 4 and 16, OTFS outperforms OFDM, while higher orders show crossover behavior and trade-offs.

IV. RESULTS

  • The results examine modulation order, mobility scenarios, number of multipath components, and equalization technique.Source: §IV · p. 4

A. Modulation Order

  • OTFS outperforms OFDM at modulation orders 4 and 16, with the performance gap increasing at higher SNR and lower modulation order.Fig. 3 compares BER across modulation orders using LMMSE, P = 5, and 380 km/h.Source: A. Modulation Order · p. 4, Figure 3 · p. 4
  • At modulation order 64, OTFS outperforms OFDM only above approximately 26 dB SNR.The results also report crossover performance points and trade-offs at orders 64 and 128.Source: A. Modulation Order · p. 4

B. Mobility Scenarios

The comparisons show OTFS robustness increasing relative to OFDM with mobility and multipath, while OTFS-MP improves on OTFS-LMMSE at low and mid-range SNR. The results cover mobility, path count, and equalization.

B. Mobility Scenarios

  • At 380 km/h, OTFS BER is almost an order of magnitude lower than OFDM at 15 dB SNR and about four orders lower at 25 dB.The performance gap increases with mobility; OTFS also outperforms OFDM in the static channel.Source: Figure 4 · p. 4, B. Mobility Scenarios · p. 4

C. Number of Multipaths Components

D. Equalizers

  • OTFS-MP outperforms OTFS-LMMSE below 21 dB SNR, with MP leveraging channel sparsity and iterative detection at greater computational complexity.Source: Figure 6 · p. 5, D. Equalizers · p. 5

V. CONCLUSIONS

The paper compares OTFS and OFDM across modulation order, mobility, multipath, and equalization. OTFS outperforms OFDM in high mobility and multipath scenarios, especially at lower modulation orders, while MP improves OTFS detection at low and mid SNR.

V. CONCLUSIONS

  • The unified evaluation compares modulation order, mobility, multipath components, and equalization to clarify where each scheme and configuration excels.Source: §V · p. 5
  • OTFS outperforms OFDM in high mobility and multipath scenarios, mainly at modulation orders 4 and 16; at orders 64 and 128, its advantage is less significant and occurs only at high SNR.MP also gains over OTFS-LMMSE at 0–21 dB SNR.Source: §V · p. 5
  • Future work includes evaluating OFDM variants, other equalization techniques, PAPR, and computational cost.Source: §V · p. 5
Loading 2609.11623v1…