R1-2601895
discussion
Discussions on 6G Waveforms
Summary
This document from Lekha Wireless Solutions presents 10 proposals on 6G waveform enhancements for both downlink and uplink. It focuses on DFT-s-OFDM adoption for downlink power efficiency in specific use cases, evaluation methodologies using realistic PA models with Net Gain metrics, and uplink PAPR reduction techniques with emphasis on the hybrid π/2-BPSK + Tone Reservation scheme.
Position
Lekha Wireless Solutions proposes supporting DFT-s-OFDM for downlink in coverage-limited cells and small-cell base stations for power-efficient transmission, while CP-OFDM remains preferred for general DL due to multi-carrier capabilities. They propose evaluating transmit power gain under realistic PA constraints (Rapp model, memory polynomial model) with ACLR limits of 45 dB for FR1 and EVM limits per modulation (QPSK < 17.5%, 16QAM < 12.5%, 64QAM < 8%, 256QAM < 3.5%), using the Net Gain metric defined as Tx power gain minus link loss at 10% BLER. For CP-OFDM downlink, they present a comparative evaluation showing SLM achieves a moderate PAPR decrease from 9.47 dB to 8.62 dB while TR provides a smaller reduction to 8.94 dB, both preserving BER performance at 9.2 dB SNR for 10⁻³ BER. For DFT-s-OFDM uplink, they demonstrate that the hybrid π/2-BPSK + TR scheme achieves the lowest PAPR (2.55 dB mean vs. 5.32 dB baseline QPSK) with improved BER performance (0.5-1 dB SNR at 10⁻³ vs. 3.5-4.0 dB baseline), and they recommend UL DFT-s-OFDM deployments up to Rank = 2 for balancing throughput, PA efficiency, and link performance.
Key proposals
- Proposal 1 (Sec 1): DFT-s-OFDM has benefits in terms of PAPR and receiver performance. Due to the multi-carrier capabilities of CP-OFDM, it is more preferred in DL. However, there are use-cases like coverage-limited cells, small-cell BSs etc., where DFT-s-OFDM can help in power efficient transmission.
- Proposal 2 (Sec 2.1): Transmit power gain needs to be calculated under realistic PA constraints while ensuring compliance with ACLR and EVM limits. Complementary metrics—such as SNR degradation, effective occupied bandwidth, and PA efficiency can also be analysed to provide a comprehensive assessment of waveform linearity and spectral behaviour.
- Proposal 3 (Sec 3): Plain OFDM works well in DL but has some limitations such as high PAPR, sensitivity to phase noise and Doppler, out-of-band emissions, and limited flexibility to mixed numerologies. Hence, enhancements and alternative OFDM-based schemes need to be explored including windowing and filtering, sub-band filtering, cyclic prefix-based enhancements, precoding and MIMO enhancements, and DFT-s-OFDM.
- Proposal 4 (Sec 3.1): Downlink transmit power gain needs to be calculated under realistic PA constraints while ensuring compliance with ACLR and EVM limits. Evaluation is based on the Net Gain metric (Tx power gain – link loss at 10% BLER) using a realistic base station PA model, complemented by secondary metrics such as PA efficiency, ACLR, EVM, network energy savings, and occupied bandwidth.
- Proposal 5 (Sec 3.2): By maintaining BER performance that is almost the same as the baseline and obtaining a moderate PAPR decrease, SLM offers an equitable trade-off. Conversely, TR retains excellent BER behavior while producing only a slight boost in PAPR.
- Proposal 6 (Sec 4): UE transmit power is limited; hence uplink need waveforms with low PAPR for efficient PA usage, robust to mobility, CFO, and Doppler and need spectral efficiency and coexistence with other UL transmissions. Enhancements to current waveforms can be done using windowing and filtering techniques, CP length variations, SC-FDMA variants, and DFT-s-OFDM.
- Proposal 7 (Sec 4.1): Due to restricted power requirements in uplink, DFT-s-OFDM was chosen for LTE UL instead of plain OFDM because it already offers lower PAPR. Still, additional PAPR reduction techniques are desired. Some good PAPR reduction techniques include DFT precoding, sub-carrier mapping, tone reservation or injection, clipping and filtering, and precoding-based techniques.
- Proposal 8 (Sec 4.2): TR provides measurable PAPR reduction while causing minimal degradation to BER compared to the baseline DFT-s-OFDM. Spectrum truncation-based methods in conjunction with TR do not give much PAPR reduction without causing severe BER degradation. Zadoff–Chu precoding yields modest PAPR improvement with increased processing complexity, and its hybridization with tone reservation offers limited additional benefit. In contrast, π/2-BPSK modulation significantly reduces PAPR without affecting BER performance. The hybrid π/2-BPSK + TR scheme achieves the lowest PAPR among all evaluated DFT-s-OFDM techniques while preserving BER transparency.
- Proposal 9 (Sec 4.3): The π/2-BPSK+TR hybrid provides additional UL PAPR reduction in DFT-s-OFDM without significant receiver changes, BER degradation, or waveform modification, enabling improved UE power efficiency.
- Proposal 10 (Sec 4.4): UL DFT-s-OFDM with Rank > 1 offers a practical means to extend uplink throughput while preserving low PAPR and high PA efficiency. Practical deployments are recommended up to Rank = 2, balancing efficiency, complexity, and link performance relative to CP-OFDM with higher ranks.