CBL - Campus del Baix Llobregat

Projecte llegit

Títol: Performance comparison of OTFS vs OFDM wireless signals in high-mobility environments

Estudiants que han llegit aquest projecte:

Director/a: ALONSO ZÁRATE, LUIS

Departament: TSC

Títol: Performance comparison of OTFS vs OFDM wireless signals in high-mobility environments

Data inici oferta: 25-06-2025     Data finalització oferta: 25-02-2026


Estudis d'assignació del projecte:
    GR ENG SIS TELECOMUN
    GR ENG SIST AEROESP
    GR ENG TELEMÀTICA
Tipus: Individual
 
Lloc de realització: EETAC
 
Paraules clau:
Wireless communications, OFDM, OTFS, Doppler, Mobility, Performance
 
Descripció del contingut i pla d'activitats:
This bachelor's thesis project focuses on the simulation and comparative performance analysis of Orthogonal Time Frequency Space (OTFS) and Orthogonal Frequency Division Multiplexing (OFDM) modulation schemes in high-mobility communication environments. As wireless communication systems progress toward 6G, ensuring reliable performance under rapidly varying channel conditions has become a crucial challenge. OFDM, while dominant in 4G and 5G networks, suffers from performance degradation in scenarios involving high Doppler spread-such as vehicular, UAV, and satellite communication-due to the loss of subcarrier orthogonality. In contrast, OTFS operates in the delay-Doppler domain and has shown strong potential for robustness in time-varying channels. This project will provide students an opportunity to gain hands-on experience with both modulation schemes and explore their strengths and weaknesses through simulation-based analysis.
The main objective of this project is to implement the baseband signal processing chains of both OFDM and OTFS in MATLAB (or optionally Python). The student will simulate and compare their Bit Error Rate (BER) and Spectral Efficiency under different levels of Doppler spread. To make the simulations realistic, the project will incorporate mobility models that reflect real-world use cases. For instance, the student may simulate scenarios such as a car moving through an urban area at 80 km/h, a UAV flying at 150 km/h, or a satellite communication link involving extremely high Doppler shifts. These channel conditions can be modeled using standard Rayleigh or Rician fading models with Doppler components. Additionally, vehicular mobility traces from platforms such as SUMO (Simulation of Urban Mobility) or synthetic mobility data can be used to enrich the simulation with more practical dynamics.
The project will also involve plotting performance metrics-such as BER versus Doppler frequency-to evaluate and visualize the impact of mobility on each waveform. The student will write a detailed report discussing the observed trade-offs in performance, computational complexity, and robustness for each waveform. Emphasis will be placed on interpreting the results in a real-world context, highlighting which waveform performs better under which type of mobility scenario and why. This hands-on project is highly relevant for students interested in wireless communication, signal processing, and emerging 6G technologies. By the end of the thesis, the student will have built a foundational understanding of modern modulation techniques and their applicability in next-generation high-mobility networks.
To ensure the accuracy and relevance of the simulation results, students will study and integrate realistic channel models that reflect the impairments experienced in dynamic wireless environments. These include modeling Doppler spread, multipath fading, and time-selective channels, which are particularly detrimental to OFDM's performance. For OTFS, the student will explore how delay-Doppler domain transforms such as the Inverse Symplectic Finite Fourier Transform (ISFFT) and Symplectic Finite Fourier Transform (SFFT) help maintain symbol integrity by spreading information across the entire time-frequency grid. As part of the analysis, students will compare how the two waveforms perform in terms of diversity gain, resilience to time variation, and latency impact. The student will be encouraged to document not just quantitative results (like BER curves), but also qualitative observations-such as implementation complexity, memory usage, and sensitivity to estimation errors. Ultimately, the project will serve as a comprehensive simulation-based feasibility study, helping inform future waveform selection decisions in the context of high-mobility 6G and ISAC systems.
 
Overview (resum en anglès):
This thesis investigates the suitability of Orthogonal Time Frequency Space (OTFS) as a waveform candidate for 5G-Advanced and beyond, by benchmarking it against conventional Orthogonal Frequency Division Multiplexing (OFDM) under high-mobility doubly selective channels. While OFDM remains the baseline in current cellular systems, its sensitivity to Doppler-induced inter-carrier interference motivates the study of alternatives that are inherently more robust in time-varying propagation conditions.

A MATLAB-based simulation framework is developed to ensure a fair, like-for-like comparison between both schemes. Multiple mobility scenarios are considered (from 0 up to 540 km/h) at a 3.5 GHz carrier, using the same time-varying multipath channel realization per frame for both links. OFDM is implemented with comb pilots, common phase error compensation and a one-tap MMSE equalizer in the time-frequency domain, whereas OTFS employs a pilot-only preamble for delay-Doppler channel estimation and a 2D MMSE equalizer. The study evaluates uncoded and coded performance using two forward error correction options-NR-LDPC (5G Toolbox) and Turbo codes (Communications Toolbox)-and reports pre-decoding BER, transport-block BER and BLER. In addition, effective spectral efficiency is computed by accounting for coding rate, overhead (cyclic prefix, pilots/preamble) and residual block errors, enabling a joint reliability-throughput assessment.

Results show that at low to moderate mobility (up to approximately 120 km/h) OFDM and OTFS exhibit similar reliability, with only minor differences at high SNR. As mobility increases, the performance gap widens and OTFS becomes increasingly advantageous, particularly in high-speed and/or high-SNR regimes where OFDM degradation due to Doppler effects is more pronounced. Building on this observation, the thesis also explores a dynamic waveform selection approach, where the transmitter chooses OFDM or OTFS based on measured channel conditions (e.g., SNR and mobility proxies such as Doppler spread or coherence time) to maximize effective throughput under a BLER constraint.

Overall, the study supports OTFS as a compelling option for high-mobility scenarios, while confirming OFDM as a competitive solution at conventional speeds. Future work includes extending the framework to higher-order modulations, improved channel estimation and receiver architectures, and a more systematic optimization of switching thresholds for adaptive OFDM/OTFS operation.

© CBLTIC Campus del Baix Llobregat - UPC