Chapter Summary
Chapter Summary
Key Points
- 1.
6G's 1000-km/h mobility target exceeds OFDM's ceiling. Enhanced OFDM numerologies can push to 500 km/h at a rate cost; OTFS handles arbitrary mobility natively. This structural limit of OFDM is the core case for OTFS in 6G.
- 2.
Dual-waveform 6G is the likely architecture. Neither OFDM nor OTFS dominates all scenarios. Dual-waveform scheduler achieves of the best-per-UE choice. OFDM for low-mobility (70-80% of use cases), OTFS for high-mobility, NTN, ISAC (20-30%).
- 3.
OTFS pays 2 dB PAPR cost over DFT-s-OFDM. Matters for UE mmWave uplink. Absorbed by OTFS's mobility + reliability gain in vehicular/LEO scenarios. Low-mobility: DFT-s-OFDM wins.
- 4.
MDMA multiplies user capacity by ~64×. DD + spatial domains add orthogonal dimensions beyond OFDMA's time-frequency. Enables 6G's -devices/km² target. OTFS is the enabling waveform.
- 5.
OTFS is 1.5× compute of 5G NR. Modern silicon (3nm, 2nm) easily absorbs this: ~20% area, ~30% power. Chip cost +~500 per BS — negligible in system cost.
- 6.
Energy: OTFS is more efficient under mobility, less under static. PAPR penalty dominates at static; retransmission savings dominate at mobility. Waveform selection (dual-mode) optimizes per UE.
- 7.
IPR: Cohere FRAND + academic alternatives. License cost ~0.5-1% of chip revenue. Comparable to LDPC. Not a blocker. CommIT contributions (Chapters 17-18) provide IPR-free academic anchors.
- 8.
Standardization timeline: Rel. 20 (2026-2028) Study Item, Rel. 21 (2028-2030) Work Item, first commercial 2030+. Mass deployment 2032+. 1-2 year risk of slippage to Rel. 22 if vendor consensus falters.
- 9.
Ecosystem health is strong (2026). Qualcomm committed, 20+ academic groups active, industry probability of 6G adoption. CommIT's Mohammadi-Caire and Buzzi-Caire contributions (cell-free, LEO) are the quantitative anchors driving adoption.
Looking Ahead
Chapter 20 dives into pulse shaping for OTFS — the practical design of transmit/receive filters under the bi-orthogonality constraint. Chapter 21 takes up AI/ML integration into OTFS receivers: deep learning detection, learned pilots, and model- based deep unfolding. Chapter 22 closes the book with open problems: optimal fractional-Doppler pilots, OTFS at terahertz, low-resolution ADCs, and the continuing OTFS-vs-enhanced-OFDM standardization debate.