Dual Threshold Self-Corrected Minimum Sum Algorithm for 5G LDPC Decoders
Round 1
Reviewer 1 Report
Although the main topic of the article is interesting, it lacks a series of key points to be able to consider the contribution as valid and to evaluate its performance within the state of the art.
- From the hardware point of view, using Vivado HLS and not an HDL language is questionable. But even worst is not providing the area and power results and making a serious comparison with the state-of-the-art. There are tons of LDPC decoder implementations in FPGAs and ASIC, and a rigorous comparison with the most efficient ones should be provided.
- In that sense, the paper has a lot of important missing references.
- There are multiple errors in the edition of the equations, and the whole paper should be revised again from a formal point of view.
- There is not a clear explanation of how thresholds are obtained, how can be obtained with other codes.
- There is not a clear flow diagram of the algorithm. What happens with the messages that are not reliable or inside the thresholds?
I think that all these points need to be clarified before the paper gets published.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
1. The author manuscript said “The algorithms of BP, MS, NMS, SCMS, DT-SCMS are simulated with the parameters setting as Table 2.”
Please write the references at the BP, MS, NMS, SCMS.
2. The author manuscript said “Table 4. Computational complexity of decoding algorithms.”
Please write the reference of BP, MS, NMS, SCMS.
3. Please give the example of the th1 and th2.
4. Please give an example of the variable dual threshold.
5. Please write the references of the (2)(3)(4)(5)(6)(7)(8)(9)(10)(16)(17).
6. The author manuscript said the equation (14).
Why is the dual threshold good better than SCMS?
7. Please write the references of Fig. 2 and Fig. 3.
8. Please show the equation of the probability likelihood ratio message L(Pi ).
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
Generally speaking, authors have presented us a solid work with significant novelty and contribution. Since 5G channel coding plays a critical role in the speeding up the deployment of the next-generation wireless networks, authors have chosen a timely topic in this paper to seek for breakthroughs by proposing an LDPC decoder atrechitecutre with dual threshold self-corrected minmum sum algorithm. The formulation, numerical analysis, and results visualization are easy to follow and understand. There are several points authors may think about to further improve the quality: first, some grammatical errors and confusing expressions need to be well handled. Second, in the introduction, a bit more elaboration could be done on the 5G NR, therefore, [1] is recommended for considerations.
[1] Y. Huo, X. Dong, W. Xu and M. Yuen, "Enabling Multi-Functional 5G and Beyond User Equipment: A Survey and Tutorial," in IEEE Access, vol. 7, pp. 116975-117008, 2019, doi: 10.1109/ACCESS.2019.2936291.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
1. It is good for the references, but it is not enough.
Please write the references in Figure 3, Figure 4, and Table 3.
2. Please give the equation of the adjustment factors of θ1 and θ2.
3. Please give the example of the th1 and th2 in the author manuscript.
4. Please give an example of the variable dual threshold in the author
manuscript.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 3
Reviewer 2 Report
1. The answer is not good enough.
The author manuscript said “So, we just give the values of the adjustment factors of θ1 and θ2, and the thresholds of th1 and th2 are not fixed in the iterations.”
I just want to adjust the θ1 and θ2.
How to do that automatically?
Please give the equation of the adjustment factors of θ1 and θ2.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
