Implementation Efficiency of Falcon Digital Signature Scheme on Arty-7 XC7A35T Board

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In abstract, it is mentioned that “Given the achieved results in terms of execution time and energy consumption, the pro-22 posed device and software may be suitable for certain applications that do not require 23 stringent performance and energy constraints”

However, falcon algorithm could be efficient solution and offer higher performance in comparison with other post-quantum algorithms such as SPHINCS+ algorithm. Therefore, the author should clarify this issue and avoid confusion.  

 

To verify the performance of Falcon algorithm, authors may consider to conduct comparisons with other post-quantum crypto algorithms alternatives such as Dilithium and SPHINCS+.

Furthermore, authors can conduct comparison with classical digital signature algorithms, such as Elliptic Curve Cryptography (ECC) and RSA. This will provide useful insights about the differences between post-quantum and classical cryptosystems in respect to performance and resources usage or utilization. Such insights will serve as a useful reference for those looking to transition from classical cryptosystems to post-quantum alternatives, ensuring long-term security for telecommunications systems.

References should be cited within the text in sequential order. For example, reference [14] is cited before [8].

Figures 1 and 2 should be mentioned first in the text before placing them in the paper. If figures are extracted from other sources, the must be appropriate citations for those sources. This comment applies for all figures in the paper.

The “Algorithm 1 (signature authentication” is mentioned in the literature review section without proper citation. If it is taken from other sources, provide the source, or you may move the algorithm to other sections if it is created in this study.

In the comparison results provided in table 5 in page 17, add a paragraph to highlight the key differences between the proposed Falcon algorithm implementation and alternatives proposed in the research studies [25], [24], and [7].

The conclusion section can be expanded to include a paragraph highlighting future research studies.

Author Response

Comments 1: In abstract, it is mentioned that “Given the achieved results in terms of execution time and energy consumption, the pro-22 posed device and software may be suitable for certain applications that do not require 23 stringent performance and energy constraints”

Response 1: Authors appreciate the reviewer's concern; this is considered in the corrected version.

1. Design Trade-off: Our primary objective was to demonstrate the feasibility of implementing a fully functional Falcon system on a low-cost FPGA platform (Xilinx Artix-7). To minimize hardware cost and avoid the complexity associated with advanced FPGA configurations, we adopted simplified hardware/software co-design and employed low-speed UART communication for data transfer.
2. Clarification on “Non-stringent Performance”: The use of a low-cost FPGA and UART interface introduces a deliberate speed trade-off. The relatively slow Key Generation time (1516.2 ms) primarily results from UART I/O bottlenecks, rather than Falcon’s computational core. Consequently, the statement that our prototype is suitable for applications without stringent performance requirements reflects this I/O constraint and our predefined hardware resource budget.

Comments 2:

However, falcon algorithm could be efficient solution and offer higher performance in comparison with other post-quantum algorithms such as SPHINCS+ algorithm. Therefore, the author should clarify this issue and avoid confusion.

To verify the performance of Falcon algorithm, authors may consider to conduct comparisons with other post-quantum crypto algorithms alternatives such as Dilithium and SPHINCS+.

Furthermore, authors can conduct comparison with classical digital signature algorithms, such as Elliptic Curve Cryptography (ECC) and RSA. This will provide useful insights about the differences between post-quantum and classical cryptosystems in respect to performance and resources usage or utilization. Such insights will serve as a useful reference for those looking to transition from classical cryptosystems to postquantum alternatives, ensuring long-term security for telecommunications systems.

Response 2: Authors appreciate the reviewer's concern; this is considered in the corrected version.

We have expanded Table 6 to include a comprehensive comparison across multiple PQC algorithms (Falcon, Dilithium, SPHINCS+) and classical schemes (ECC, RSA). This table highlights the resource utilization and execution time differences between lattice-based and hash-based approaches, offering a clearer performance context.

Comments 3: References should be cited within the text in sequential order. For example, reference [14] is cited before [8].

Response 3: Authors appreciate the reviewer's concern; this is considered in the corrected version, and we improved the quality of the paper.

Comments 4: Figures 1 and 2 should be mentioned first in the text before placing them in the paper. If figures are extracted from other sources, the must be appropriate citations for those sources. This comment applies for all figures in the paper.

Response 4: Authors appreciate the reviewer's concern. We corrected mistakes and we improved the quality of the paper.

Comments 5: The “Algorithm 1 (signature authentication” is mentioned in the literature review section without proper citation. If it is taken from other sources, provide the source, or you may move the algorithm to other sections if it is created in this study.

Response 5: Authors appreciate the reviewer's concern; this is considered in the corrected version.

Comments 6: In the comparison results provided in table 5 in page 17, add a paragraph to highlight the key differences between the proposed Falcon algorithm implementation and alternatives proposed in the research studies [25], [24], and [7].

Response 6: Authors appreciate the reviewer's concern; this is considered in the corrected version.

Comments 7: The conclusion section can be expanded to include a paragraph highlighting future research studies.

Response 7: Authors appreciate the reviewer's concern; this is considered in the corrected version.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This paper investigates the implementation of the Falcon post-quantum digital signature algorithm on a low-cost FPGA platform (Arty-7 XC7A35T) and proposes a hardware/software co-design approach. In this design, several modules of the Falcon algorithm (TRNG, modular addition, and signature verification) are migrated to the hardware side, while computationally intensive components remain executed on the software side.

1. The TRNG output is not entropy-conditioned, which may cause bias or correlation, affecting key security.

2. The randomness of the TRNG module has not been validated using NIST SP 800-22 / 90B tests.

3. It is recommended to provide entropy analysis or statistical test results to demonstrate the quality and security of the random numbers.

4. It is suggested to unify symbol naming and add legends and data flow descriptions.

5. To further extend the application of the post-quantum signature algorithm in multimedia and embedded security scenarios, it is recommended to refer to "Encrypt a Story: A Video Segment Encryption Method Based on the Discrete Sinusoidal Memristive Rulkov Neuron," which employs a memristive neuron model for efficient video segment encryption. Its chaotic mapping and hardware-friendly features can provide useful insights for extending Falcon signatures to secure data transmission.

6. To enhance the synergy between data compression and security protection, the authors may cite "Image Privacy Protection Scheme Based on High-Quality Reconstruction DCT Compression and Nonlinear Dynamics." This study combines nonlinear dynamics with high-quality DCT reconstruction for privacy protection, offering valuable guidance for optimizing the hardware-based secure signature system discussed in this paper.

7. In the field of digital signature anti-forensics and robustness, it is suggested to cite "A GAN-based Anti-forensics Method by Modifying the Quantization Table in JPEG Header File." This work introduces a generative adversarial network to modify JPEG quantization tables for anti-forensic defense, providing meaningful reference ideas for improving Falcon signature systems in terms of forgery resistance and robust verification.

Author Response

Comments 1:

This paper investigates the implementation of the Falcon post-quantum digital signature algorithm on a low-cost FPGA platform (Arty-7 XC7A35T) and proposes a hardware/software co-design approach. In this design, several modules of the Falcon algorithm (TRNG, modular addition, and signature verification) are migrated to the hardware side, while computationally intensive components remain executed on the software side.

The TRNG output is not entropy-conditioned, which may cause bias or correlation, affecting key security.

The randomness of the TRNG module has not been validated using NIST SP 800- 22/90B tests.

It is recommended to provide entropy analysis or statistical test results to demonstrate the quality and security of the random numbers.

Response 1: Authors appreciate the reviewer's concern; This was described in section 2.2, page 7

Comments 2: It is suggested to unify symbol naming and add legends and data flow descriptions.

Response 2: Authors appreciate the reviewer's concern; this is considered in the corrected version.

Comments 3:

To further extend the application of the post-quantum signature algorithm in multimedia and embedded security scenarios, it is recommended to refer to "Encrypt a Story: A Video Segment Encryption Method Based on the Discrete Sinusoidal Memristive Rulkov Neuron," which employs a memristive neuron model for efficient video segment encryption. Its chaotic mapping and hardware-friendly features can provide useful insights for extending Falcon signatures to secure data transmission.

To enhance the synergy between data compression and security protection, the authors may cite "Image Privacy Protection Scheme Based on High-Quality Reconstruction DCT Compression and Nonlinear Dynamics." This study combines nonlinear dynamics with high-quality DCT reconstruction for privacy protection, offering valuable guidance for optimizing the hardware-based secure signature system discussed in this paper.

In the field of digital signature anti-forensics and robustness, it is suggested to cite "A GAN-based Anti-forensics Method by Modifying the Quantization Table in JPEG Header File." This work introduces a generative adversarial network to modify JPEG quantization tables for anti-forensic defense, providing meaningful reference ideas for improving Falcon signature systems in terms of forgery resistance and robust verification.

Response 3: Authors appreciate the reviewer's concern. These will be the focus of the Future Work section, which has been expanded and revised as recommended.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The proposed research implements Falcon, a quantum-resistant cryptography solution developed by IBM. It combines hardware and software, rather than traditional software approaches, to achieve short authentication times. However, key generation time, signing time, and verification time are all found to be inferior to those of existing research. Nevertheless, the proposed hardware solution appears to have performed well, and it is necessary to implement various digital signatures in preparation for the quantum era.

Author Response

Comments 1: The proposed research implements Falcon, a quantum-resistant cryptography solution developed by IBM. It combines hardware and software, rather than traditional software approaches, to achieve short authentication times. However, key generation time, signing time, and verification time are all found to be inferior to those of existing research. Nevertheless, the proposed hardware solution appears to have performed well, and it is necessary to implement various digital signatures in preparation for the quantum era.

Response 1: Thank you for your valuable comment. These will be the focus of the Future Work section, which has been expanded and revised as recommended.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for your effort. The majority of recommendations have been addressed. However, there still some minor issues need to be addressed, as follows:

In page 2, add an introductory paragraph before section 2.2.

On page 4, please mention Figure 2 in the text before inserting it.

On page 8, please mention Figure 4 in the text before inserting it.

On page 10, please mention Figure 7 in the text before inserting it.

On page 11, please mention table 1 in the text before inserting it.

On page 12, please mention table 2 in the text before inserting it.

On page 13, please mention table 3 in the text before inserting it.

On page 14, please mention Figure 14 in the text before inserting it.

On page 15, please mention Figure 15 in the text before inserting it.

On page 17, please mention Figure 17 in the text before inserting it.

On page 17, please mention Figure 18 in the text before inserting it.

On page 18, please mention table 6 in the text before inserting it.

 

Author Response

Comments 1: In page 2, add an introductory paragraph before section 2.2.

Response 1: We appreciate the reviewer views. This is correct, The authors added an additional paragraph to page 2 between sections 2 and 2.1.

 Comments 2: On page 4, please mention Figure 2 in the text before inserting it.

Response 2: We appreciate the reviewer views. This is correct, the authors have corrected in the revised paper.

Comments 3: On page 8, please mention Figure 4 in the text before inserting it.

Response 3: We appreciate the reviewer views. This is correct, the authors have corrected in the revised paper.

Comments 4: On page 10, please mention Figure 7 in the text before inserting it.

Response 4: We appreciate the reviewer views. This is correct, the authors have corrected in the revised paper.

Comments 5: On page 11, please mention table 1 in the text before inserting it.

Response 5: We appreciate the reviewer views. This is correct, the authors have corrected in the revised paper.

Comments 6: On page 12, please mention table 2 in the text before inserting it.

Response 6: We appreciate the reviewer views. This is correct, the authors have corrected in the revised paper.

Comments 7: On page 13, please mention table 3 in the text before inserting it.

Response 7: We appreciate the reviewer views. This is correct, the authors have corrected in the revised paper.

Comments 8: On page 14, please mention Figure 14 in the text before inserting it.

Response 8: We appreciate the reviewer views. This is correct, the authors have corrected in the revised paper.

Comments 9: On page 15, please mention Figure 15 in the text before inserting it.

Response 9: We appreciate the reviewer views. This is correct, the authors have corrected in the revised paper.

Comments 10: On page 17, please mention Figure 17 in the text before inserting it.

Response 10: We appreciate the reviewer views. This is correct, the authors have corrected in the revised paper.

Comments 11: On page 17, please mention Figure 18 in the text before inserting it.

Response 11: We appreciate the reviewer views. This is correct, the authors have corrected in the revised paper.

Comments 12: On page 18, please mention table 6 in the text before inserting it.

Response 12: We appreciate the reviewer views. This is correct, the authors have corrected in the revised paper.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The authors's revisions satisfactorily resolve the raised issues, and the manuscript now meets the publication standards. I recommend that it be accepted.

Author Response

Comments 1: The authors's revisions satisfactorily resolve the raised issues, and the manuscript now meets the publication standards. I recommend that it be accepted.

Response 1: Thank you for your valuable comment.

Author Response File: Author Response.pdf