- Rating
- 2(+1)
- Confidence
- 4
- Informativeness
- 1
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
This paper proposes G-Wave, an end-to-end neural transceiver that directly maps LDPC-coded bits to time-domain IQ waveforms and maps the received waveform back to coded-bit logits, without relying on a OFDM resource grid. G-Wave shows constantly better result compared to DeepOFDM, DeepOFW and all the other baselines.
Critical strengths
G-Wave consistently outperforms all baselines, even under perfect CSI. The paper provides good analyses of what the learned waveform actually does. Fig. 6 clearly shows it for several cases. The authors also honestly report cases when spectral adaptation is not correct in the appendix.
Critical weaknesses
The paper has several weaknesses. Fairness to the baselines, an overly simple setting, limited structural novelty in the model, and the use of BER as the only metric. I detail these below in the questions. My biggest question is this: if we only consider stationary users, why should we not just use OFDM? Where does all of G-Wave's gain actually come from?
Questions for authors
- Most critically, the "No conditioning" variant of G-Wave (9.22e-3) performs on par with DeepOFDM (9.52e-3). The proposed method should therefore be compared against OFDM with water-filling based on CSI, or at least on estimated CSI. In such a comparison, I suspect the benefit of the framework itself would be marginal. I am not claiming that there is no effect. However, if the improvement comes mainly from the conditioning itself, then most of the contribution cannot be attributed to the proposed framework.
- The experimental setting is overly simplistic. Single-user SISO, CDL-A~E channels, and a static channel at 0 m/s. The paper claims SOTA using only standardized LOS/NLOS profiles, without considering site-specific propagation (e.g., via ray tracing). This limits the practical applicability of the results. In particular, user mobility is crucial for this line of work, as emphasized in DeepOFDM, and the complete absence of mobility is a critical limitation.
- The delay spread used for training is 0.1–0.6 μs, which seems too long compared to typical indoor settings.
- Although a Transformer is used, isn't the model essentially trained in a supervised manner? It is neither a diffusion model nor even a GAN, so it is unclear in what sense the method is "generative", in what sense it learns a distribution. It is certainly a neural TX/RX model, but I do not think it qualifies as a generative model.
- DeepOFW was originally designed for PAPR reduction, so comparing against it solely in terms of BER seems unfair to DeepOFW.
- As I understand it, G-Wave's gain comes from channel-adaptive power allocation across frequency. However, this could also be achieved within DeepOFDM, and OFDM-based approaches that operate on a fine-grained per-subcarrier basis could likely achieve it even more explicitly. If gridity was the problem, ablation could be done with it. Yet no ablation addresses this. In other words, even if G-Wave is state of the art among the compared methods, it is not evident that the improvement stems from properties unique to G-Wave.
- The CP/pilot overhead is present only on the OFDM side, which makes the comparison unfair.
- Only the arithmetic mean of BER over the evaluated Eb/N0 points is reported. Neither BLER nor the dB gap at a target BER is provided.
- The architecture combines VAR-style multiscale modeling [A] with DiT's AdaLN-Zero conditioning [B], and training is an end-to-end AE with a BCE loss [C], which is not particularly novel. Compared with [A][B][C], there is nothing architecturally distinctive. This is a significant weakness for an ICLR submission. Had the same paper been submitted to a communications venue, this might not have been a major concern. References [A] K. Tian, Y. Jiang, Z. Yuan, B. Peng, and L. Wang. Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction. NeurIPS, 2024. [B] W. Peebles and S. Xie. Scalable Diffusion Models with Transformers. ICCV, 2023. [C] T. O'Shea and J. Hoydis. An Introduction to Deep Learning for the Physical Layer. IEEE Trans. Cognitive Communications and Networking, 3(4):563–575, 2017.
6 comments
Responses for Q3 to Q7
Q3: We use the ()–(,) delay-spread range during training to expose G-Wave to more challenging, highly frequency-selective channels and better explore its adaptation capability. For indoor evaluation, we follow 3GPP TR 38.901 [2] and use the more typical ()–() ns range. Despite being outside the training distribution, G-Wave still achieves the lowest mean BER under both perfect and estimated CSI, demonstrating good generalization to typical indoor channels.
Q4: G-Wave connects the bit space and waveform space through a learned latent representation. The tokenizer first encodes the payload bits into latent features. The channel condition then guides the transformation of these features, and the synthesis head converts them into a physical waveform. The bits specify the information to be transmitted, while the condition guides how that information is represented in the waveform.
For a fixed channel condition, different bit sequences produce a collection of waveforms whose distribution is induced by the bit distribution. Changing the condition changes the latent transformation and consequently the waveform distribution. Through joint training with the receiver, G-Wave learns these condition-dependent representations for reliable bit recovery, without predefined waveform labels. This learned transformation from a bit distribution, through conditioned latent representations, to physical waveforms is the mechanism underlying our use of “generative waveform”. We will make this relationship explicit in the revised manuscript.
Q5: For a fair comparison using the same training objective, we specifically trained DeepOFW without a PAPR penalty. We now also report PAPR for this baseline and G-Wave. With oversampling, G-Wave has a mean PAPR of dB, compared with dB for BCE-only DeepOFW; at a CCDF level of , the corresponding values are and dB. Thus, the additional comparison reports peak-power behavior as well as error rates, with the baseline training objective explicitly specified.
Table 4. PAPR with oversampling (dB).
| Method | Mean | CCDF | CCDF |
|---|---|---|---|
| G-Wave | 7.32 | 10.16 | 10.76 |
| DeepOFW, BCE only | 8.73 | 10.64 | 11.52 |
| DeepOFDM, with PAPR penalty | 10.34 | 11.59 | 11.82 |
| DeepOFDM, without PAPR penalty | 10.58 | 12.56 | 13.34 |
| OFDM (QPSK) | 8.51 | 10.33 | 10.99 |
Q6: We clarify that G-Wave’s gain does not come solely from channel-adaptive power allocation. Rather, the framework jointly learns several aspects of waveform design: (1) a flexible bit-to-waveform representation, where information is distributed across multiple waveform samples; (2) channel-adaptive power allocation, which reshapes transmitted energy according to the channel condition; and (3) channel-adaptive overhead allocation, which avoids the fixed pilot/CP overhead of conventional OFDM and can instead adjust the amount of redundancy according to the channel condition.
These effects are learned jointly through end-to-end optimization and cannot be reduced to water-filling alone. To isolate the contribution beyond channel-adaptive power allocation, we additionally evaluate OFDM with CSI-based water-filling. It still exhibits substantially higher mean BER than G-Wave under both perfect and estimated CSI (() vs. (), and () vs. (), respectively).
Regarding resource-grid constraints, the requested ablation is already included in Table 2. The “Frequency-domain I/O” and “OFDM resource-grid I/O” variants retain the same neural backbone and conditioning while progressively constraining waveform generation, yielding BERs of () and (), respectively, compared with () for full G-Wave. These results provide further evidence that G-Wave’s gain comes from the joint optimization of waveform representation, channel-adaptive power allocation, and adaptive resource overhead, rather than from power allocation alone. We will clarify this distinction more explicitly in the revised manuscript.
Q7: We clarify that the comparison uses the same number of bits, the same transmission duration, bandwidth, and the same frame-energy budget for all methods. Conventional OFDM allocates part of this fixed time-frequency resource to CP and pilots because they are required for channel estimation and equalization. G-Wave does not explicitly reserve such structures, but must instead learn the required robustness and channel adaptation end-to-end within the same transmission budget. Therefore, G-Wave is not given additional bits, time, or energy; the difference lies only in how each method utilizes the same available transmission resources.
Response to Q8
Q8: We thank the reviewer for this valuable suggestion. We have conducted additional BLER evaluations and provide a more detailed analysis of the SNR gains at target error rates. At an information-bit BER of , G-Wave achieves gains of approximately dB and dB over DeepOFDM under perfect and estimated CSI, respectively. Furthermore, under perfect CSI, G-Wave reaches BLER at dB, compared with dB for DeepOFDM, yielding a dB advantage. These additional results further demonstrate G-Wave's reliability improvements beyond arithmetic-mean BER. We will incorporate these analyses into the revised manuscript.
Table 5. Mean BLER and SNR at BLER (Exp. 1).
Perfect CSI
| Method | Mean BLER | at BLER (dB) | Gap to G-Wave (dB) |
|---|---|---|---|
| G-Wave | 7.83 | 0.00 | |
| DeepOFW, BCE only | 9.03 | 1.20 | |
| DeepOFDM | 10.42 | 2.59 | |
| DeepOFDM, no conditioning | 12.24 | 4.41 | |
| DeepOFDM, BCE only | 10.06 | 2.23 | |
| OFDM-LMMSE | 12.31 | 4.48 |
Estimated CSI
| Method | Mean BLER | at BLER (dB) | Gap to G-Wave (dB) |
|---|---|---|---|
| G-Wave | 8.25 | 0.00 | |
| DeepOFW | — | ||
| DeepOFDM | 11.15 | 2.90 | |
| OFDM-LS | — |
Table 6. SNR and gaps at information-bit BER .
| Method | Perfect CSI: at BER (dB) | Perfect CSI: Gap to G-Wave (dB) | Estimated CSI: at BER (dB) | Estimated CSI: Gap to G-Wave (dB) |
|---|---|---|---|---|
| G-Wave | 12.26 | 0.00 | 12.44 | 0.00 |
| DeepOFDM | 14.40 | 2.14 | 13.93 | 1.50 |
| DeepOFW | 12.85 | 0.59 | — | |
| OFDM | 14.95 | 2.68 | — |
Responses for Q1 and Q2
Q1: We thank the reviewer for raising this point and will clarify it in the revision. G-Wave consistently outperforms DeepOFDM under matched settings, both with and without CSI conditioning. With the same CSI input, G-Wave reduces mean BER by 48.5% and 42.4% over DeepOFDM under perfect and estimated CSI, respectively. More importantly, even without any CSI conditioning, G-Wave achieves a lower BER than CSI-conditioned DeepOFDM (9.22\times10^{-3}) vs. (9.52\times10^{-3})). Therefore, the gain cannot be attributed primarily to conditioning; the proposed framework itself provides a substantial advantage.
To further isolate the benefit beyond conventional CSI-aware adaptation, we additionally compare against OFDM with CSI-based water-filling under both perfect and estimated CSI. G-Wave still consistently achieves lower BER, further confirming that its advantage extends beyond CSI conditioning or power allocation.
Table 1. Mean information-bit BER with CSI conditioning and water-filling.
G-Wave thus retains lower mean BER than CSI-based water-filling OFDM under both perfect and estimated CSI.
Q2: Prior neural communication and waveform-learning works, including DeepOFDM, have adopted similar single-user SISO settings with standardized channel models. We follow this established setup to enable a controlled and fair comparison with prior work. Moreover, 3GPP CDL-A--E models cover diverse LOS/NLOS conditions, multipath structures, and delay spreads [2], making them meaningful benchmarks for waveform design.
To further evaluate generalization beyond standardized channel models, we additionally test G-Wave, without retraining, on two unseen site-specific ray-tracing channel datasets: Sionna RT Munich [5] and DeepMIMO ASU campus [6]. G-Wave achieves the lowest mean information-bit BER on both, reducing BER by 46.9% and 49.0% relative to the strongest baselines, respectively.
Regarding mobility, although G-Wave is not explicitly trained for mobile channels, our additional evaluation shows that it maintains lower mean information-bit BER than DeepOFDM at all tested speeds up to 72 km/h under estimated CSI. At 72 km/h, G-Wave achieves a mean BER of 3.48×10−3, compared with 4.94×10−3 for DeepOFDM. We will include these results in the revised manuscript to demonstrate that the framework is not inherently restricted to static channels.
Table 2. Mean BER on unseen ray-tracing channels.
Sionna RT Munich
DeepMIMO ASU campus
Table 3. Mean information-bit BER under mobility with estimated CSI (5, 10, and 15 dB).