Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The paper provides a comprehensive review of reliability challenges and key technologies for FPGA-based reconfigurable systems used in critical applications such as aerospace and the nuclear industry. However, some minor modifications can improve it.

• Figures and tables, while mostly informative, would benefit from improved labeling and more consistent presentation.

• The discussion of "results" is clear in scope but could be organized with greater emphasis on novel findings and specific quantitative outcomes.

• The structure is logical; however, more direct comparison with recent literature would enhance scientific impact.

• Consider providing more illustrative examples or case studies for readers less familiar with the field.

Comments on the Quality of English Language

Some sentences are overly long or awkward and would benefit from professional language editing for clarity and precision. Please address minor typographical errors scattered throughout the text.

 

Author Response

Reviewer #1, Concern #1: Figures and tables, while mostly informative, would benefit from improved labeling and more consistent presentation.

Author response: Thank you very much for your question. We completely agree that improving the labeling consistency and presentation quality of figures and tables will significantly enhance the readability and professional appearance of our manuscript.

Author action: To carefully address your comment, we have implemented a systematic icon standardization process throughout the manuscript.

 

Reviewer #1, Concern #2: The structure is logical; however, more direct comparison with recent literature would enhance scientific impact.

Author response: Thank you very much for your question. We agree that enhancing direct comparisons with recent literature will significantly strengthen the scientific contribution and impact of our review article.

Author action: To carefully address your comment, we have systematically added comparisons with recent high-quality publications (2021-2025) in key technical areas。

 

 

 

Reviewer #1, Concern #3: Consider providing more illustrative examples or case studies for readers less familiar with the field.

Author response: Thank you very much for your question. We fully agree that adding illustrative examples and case studies will significantly improve the accessibility of our review for readers who are new to the field of FPGA reliability. In response to this comment, we have implemented a comprehensive case study enhancement throughout the manuscript.

Author action: To carefully address your comment, in Section 4.3 we added a detailed case study on TMR implementation in aerospace applications, providing concrete examples of triple modular redundancy deployment in satellite control systems, complete with implementation steps, resource configuration schemes, and quantitative performance metrics including fault detection rates and system recovery time. In Section 5.4, we incorporated a machine learning-based aging prediction case study set in industrial environments, demonstrating the complete workflow from feature extraction to model validation with actual accuracy rates and false positive ratios. Furthermore, in Section 6.5, we included a hardware Trojan detection case study for critical infrastructure protection, showcasing integrated detection methodologies combining static analysis and side-channel techniques with practical detection efficiency data.

 

Reviewer #1, Concern #4: Some sentences are overly long or awkward and would benefit from professional language editing for clarity and precision. Please address minor typographical errors scattered throughout the text.

Author response: Thank you very much for your question. We fully acknowledge that the clarity and precision of academic writing are essential for effective knowledge dissemination, and we appreciate the reviewer's careful attention to these details.

Author action: To carefully address your comment, we have implemented a multi-stage language enhancement process.

Reviewer 2 Report

Comments and Suggestions for Authors

Initially, it was unclear whether the manuscript was intended as a research article or a review paper. However, based on the large number of references cited and the way the manuscript is structured, I strongly believe that it should have been submitted under the review category rather than as a regular research article. Moreover, there is no transparency in the selection of the information presented, which makes it challenging to understand the purpose of this work entirely. The manuscript does not provide a transparent methodology for article selection, nor does it specify which databases were consulted to compile the reviewed literature. For these reasons, I consider that the article requires a significant restructuring and should be resubmitted, since there are critical issues in its current structure. Nevertheless, I provide my detailed observations below, hoping that they will be constructive and helpful for the authors.

The abstract should be strengthened by including quantitative information from the review process, such as the total number of articles analyzed, the publication time frame, and the main categories identified. Adding these details will provide the reader with a more straightforward overview of the scope and depth of the survey.

Authors must consider adding a paragraph at the end of the introduction that clearly states the explicit objective of this review, outlining its scope, contributions, and intended value for the research community, since the current purpose of the study is not clearly defined. In addition, the authors should highlight the added value of this review compared to other existing surveys, for example, by emphasizing whether it provides a new classification, discusses emerging trends, or offers a critical perspective on trade-offs.

The state of the art should be updated with at least 5–10 recent references (2021–2023) addressing FPGA reliability, machine learning approaches for fault prediction, and hardware Trojan defense.

The authors must consider adding comparative tables that synthesize the reviewed solutions (e.g., TMR vs. ECC vs. dynamic reconfiguration), including relevant metrics such as overhead, efficiency, and robustness.

The section on emerging trends should be further developed by explaining open challenges, such as balancing reliability and resource costs, integration with edge computing, and cybersecurity in reconfigurable environments.

Although the authors briefly mention some future directions, it would significantly strengthen the contribution to organize these points into a dedicated section on emerging trends. This section could expand the discussion of open challenges, such as balancing reliability and resource costs, integration with edge computing, and cybersecurity in reconfigurable environments, to provide a clearer perspective for future research.

The authors must include a transparent methodology for their bibliographic search. This section should explicitly specify the databases consulted (e.g., IEEE Xplore, ScienceDirect, Scopus, Web of Science, MDPI), the main keywords used (e.g., FPGA reliability, fault tolerance, dynamic reconfiguration, hardware Trojan), and the time frame considered (e.g., 2010–2023). Without this information, the review lacks transparency and reproducibility, which are essential for a high-quality survey paper.

The authors must provide a clear and transparent methodology for the literature selection process. This should explicitly describe the inclusion and exclusion criteria (e.g., restricting the survey to peer-reviewed articles in English, excluding redundant or unrelated works, and prioritizing practical applications or benchmark studies). Additionally, the process should be documented with a flow diagram (e.g., PRISMA or another recognized methodology for systematic reviews) that shows the number of articles initially identified, the filtering stages, and the final set of references included. 

The manuscript lacks a dedicated discussion section where the most commonly used techniques are critically analyzed. While several approaches (e.g., TMR, ECC, dynamic reconfiguration, aging sensors, ML-based monitoring, hardware Trojan detection) are described, there is no systematic comparison of their strengths, limitations, or practical deficiencies.

The conclusions are too brief and do not provide sufficient support for the review. They should be expanded to summarize the main findings clearly, highlight the contributions of the survey, and emphasize the open challenges and future research directions identified throughout the manuscript.

Author Response

Reviewer #2, Concern #1: Initially, it was unclear whether the manuscript was intended as a research article or a review paper. However, based on the large number of references cited and the way the manuscript is structured, I strongly believe that it should have been submitted under the review category rather than as a regular research article. Moreover, there is no transparency in the selection of the information presented, which makes it challenging to understand the purpose of this work entirely. The manuscript does not provide a transparent methodology for article selection, nor does it specify which databases were consulted to compile the reviewed literature. For these reasons, I consider that the article requires a significant restructuring and should be resubmitted, since there are critical issues in its current structure.

Author response: Thank you very much for your question. We acknowledge that the manuscript was incorrectly positioned as a research article and agree that its true nature is a comprehensive review paper. We appreciate the reviewer's clear guidance on the necessary structural improvements.

Author action: We thank the reviewer for this essential feedback that has guided the transformation of our manuscript into a proper review article. The significant restructuring has addressed the fundamental issues identified, and we are confident that the revised manuscript now makes a valuable contribution as a comprehensive review of FPGA reliability techniques. The added methodological transparency and analytical depth significantly enhance the work's scholarly value and utility to the research community.

 

 

 

Reviewer #2, Concern #2: The abstract should be strengthened by including quantitative information from the review process, such as the total number of articles analyzed, the publication time frame, and the main categories identified. Adding these details will provide the reader with a more straightforward overview of the scope and depth of the survey.

Author response: Thank you very much for your question. We completely agree that adding quantitative information will significantly enhance the abstract's ability to convey the scope and methodology of our review, providing readers with immediate clarity on the survey's comprehensiveness.

Author action: To carefully address your comment ,we have completely rewritten the abstract to incorporate the requested quantitative elements.

 

 

 

Reviewer #2, Concern #3: Authors must consider adding a paragraph at the end of the introduction that clearly states the explicit objective of this review, outlining its scope, contributions, and intended value for the research community, since the current purpose of the study is not clearly defined. In addition, the authors should highlight the added value of this review compared to other existing surveys, for example, by emphasizing whether it provides a new classification, discusses emerging trends, or offers a critical perspective on trade-offs.

Author response: Thank you very much for your question. We fully agree that a clear statement of objectives and unique contributions is essential for positioning our review within the existing literature and demonstrating its value to the research community.

Author action: To carefully address your comment, we have added a new dedicated paragraph at the end of the introduction (Section 1, Paragraph 5) that explicitly addresses all points raised by the reviewer.

 

 

 

Reviewer #2, Concern #4: The state of the art should be updated with at least 5–10 recent references (2021–2023) addressing FPGA reliability, machine learning approaches for fault prediction, and hardware Trojan defense.

Author response: Thank you very much for your question. We fully agree that incorporating recent literature is essential for maintaining the state-of-the-art currency and relevance of our review. We have systematically updated the reference list with high-quality publications from the specified timeframe.

Author action: To carefully address your comment, we have added a number of recent high-quality references (2021-2025) across the three specified technical domains, exceeding the minimum requirement to ensure comprehensive coverage.

 

 

 

Reviewer #2, Concern #5: The authors must consider adding comparative tables that synthesize the reviewed solutions (e.g., TMR vs. ECC vs. dynamic reconfiguration), including relevant metrics such as overhead, efficiency, and robustness.

Author response: Thank you very much for your question. We completely agree that comparative tables will significantly enhance the analytical depth and practical utility of our review by providing readers with a systematic framework for evaluating different technical approaches.

Author action: To carefully address your comment, we have designed and incorporated two comprehensive comparative tables​ that systematically synthesize the technical solutions discussed throughout the review.

 

 

Reviewer #2, Concern #6: The section on emerging trends should be further developed by explaining open challenges, such as balancing reliability and resource costs, integration with edge computing, and cybersecurity in reconfigurable environments.

Author response: Thank you very much for your question. We fully agree that a more thorough exploration of open challenges will significantly enhance the forward-looking value and practical relevance of our review, particularly for researchers and practitioners seeking to advance the state of the art in FPGA reliability.

Author action: To carefully address your comment, we have substantially expanded Section 8: Emerging Trends and Open Challenges​ (previously titled "Emerging Trends") to provide a detailed analysis of critical research gaps and future directions.

 

 

Reviewer #2, Concern #7: Although the authors briefly mention some future directions, it would significantly strengthen the contribution to organize these points into a dedicated section on emerging trends. This section could expand the discussion of open challenges, such as balancing reliability and resource costs, integration with edge computing, and cybersecurity in reconfigurable environments, to provide a clearer perspective for future research.

Author response: Thank you very much for your question. We completely agree that a dedicated, well-structured section on emerging trends will provide readers with a more systematic and actionable perspective on future research directions.

Author action: To carefully address your comment, we have completely reorganized and significantly expanded the future directions discussion into a new Section 8: Emerging Trends and Research Frontiers.

 

 

Reviewer #2, Concern #8: The authors must include a transparent methodology for their bibliographic search. This section should explicitly specify the databases consulted (e.g., IEEE Xplore, ScienceDirect, Scopus, Web of Science, MDPI), the main keywords used (e.g., FPGA reliability, fault tolerance, dynamic reconfiguration, hardware Trojan), and the time frame considered (e.g., 2010–2023). Without this information, the review lacks transparency and reproducibility, which are essential for a high-quality survey paper.

Author response: Thank you very much for your question. We fully agree that a transparent and reproducible literature search methodology is fundamental to the credibility and scholarly value of a review paper. We have completely addressed this concern by adding a comprehensive methodology section.

Author action: To carefully address your comment, we have added a new Section 2: Literature Review Methodology​ that provides full transparency and follows systematic review standards.

 

 

Reviewer #2, Concern #9: The authors must provide a clear and transparent methodology for the literature selection process. This should explicitly describe the inclusion and exclusion criteria (e.g., restricting the survey to peer-reviewed articles in English, excluding redundant or unrelated works, and prioritizing practical applications or benchmark studies). Additionally, the process should be documented with a flow diagram (e.g., PRISMA or another recognized methodology for systematic reviews) that shows the number of articles initially identified, the filtering stages, and the final set of references included.

Author response: Thank you very much for your question. We fully acknowledge that a well-documented selection process is essential for the credibility and reproducibility of our survey. We have comprehensively addressed this concern by implementing a rigorous methodology following PRISMA guidelines.

Author action: To carefully address your comment, we have developed and implemented a complete systematic literature review methodology that exceeds standard practices in transparency and rigor.

 

 

Reviewer #2, Concern #10: The manuscript lacks a dedicated discussion section where the most commonly used techniques are critically analyzed. While several approaches (e.g., TMR, ECC, dynamic reconfiguration, aging sensors, ML-based monitoring, hardware Trojan detection) are described, there is no systematic comparison of their strengths, limitations, or practical deficiencies.

Author response: Thank you very much for your question. We fully agree that a dedicated discussion section providing systematic critical analysis and comparative assessment is essential for transforming a descriptive review into an analytically valuable contribution. The absence of such analysis indeed limited the manuscript's scholarly impact.

Author action: To carefully address your comment, we have added a comprehensive Section 7: Comparative Analysis and Critical Discussion​ that provides the systematic critical analysis requested by the reviewer.

 

 

Reviewer #2, Concern #11: The conclusions are too brief and do not provide sufficient support for the review. They should be expanded to summarize the main findings clearly, highlight the contributions of the survey, and emphasize the open challenges and future research directions identified throughout the manuscript.

Author response: Thank you very much for your question. We fully acknowledge that the original conclusion section was insufficient in scope and depth to properly summarize the comprehensive work presented in our review. We have completely rewritten and significantly expanded the conclusion to provide a robust and meaningful synthesis of our findings.

Author action: To carefully address your comment, we have completely restructured the Conclusion​ section into a comprehensive Section 9: Conclusions and Future Perspectives​ that addresses all aspects highlighted by the reviewer.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This paper attempts to provide a survey of reliability in FPGA systems, covering the areas of fault tolerance, aging mitigation, and hardware attack defense. The topic is timely and of significant importance to the field. With logically structured content, which starts from the threat models, then digs deep into the specific methods to counter them, this paper provides a clear path for readers new to the subject.

However, the paper in its current form suffers from a significant lack of depth, clarity, and analysis, preventing it from being the comprehensive survey it aims to be. The discussion of existing work is superficial, key literatures are omitted, and the presentation has issues. It would require some effort to address these fundamental shortcomings.

Analysis and Comprehensiveness: A survey paper's primary contribution is the synthesis and critical analysis of existing work. This manuscript, however, falls short of this.

1. Most of the prior work discussed in this paper is summarized with a simple and superficial summary, rather than a useful analysis. For example, the section on intelligent fault tolerance (lines 428 onwards) mentions a series of papers [85-88] that all use evolutionary methods. The discussion of those works is basically "they use evolutionary method with balancing / tolerance for electromagnetic damages / others". What those concepts are are not delved into. Their specific approaches, effectiveness, and trade-offs are not elaborated or compared. This is insufficient for a high-quality survey.

2. The survey is not comprehensive and omits some relevant work. A notable example is the absence of work by Cong and Gururaj on application-level correctness, which represents one perspective on fault tolerance, and the whole group of similar work of soft error handling is not covered.

3. Some concepts lack proper context and citation. The STAR (Self-Test Areas) method is presented as being specific to FPGAs and cites a 2014 paper on bitstream readback. However, the cited paper has no mention of it, and this is in fact a proposal made much earlier at Bell Labs; the original work is not cited, and the context is misleading: it is not a part of existing FPGA devices.

Presentation and Clarity: The manuscript has quite a few issues in its presentation, including wording, formatting, and figure presentations.

1. Phrases such as "at home and abroad" are not appropriate for international journals, and unconventional terms like "research windfall", "so-called FPGA" should be revised. The overall text would benefit from a proofreading by a native English speaker.

2. Figure 1 is an art illustration of application areas that adds little value and would be better served by descriptive text with citations. The text mentions military applications, which should also be cited. Figure 2 does not clearly illustrate the Single Event Effects (SEE) it purports to, and the accompanying text lists several SEE types (SEU, SET, SEFI, etc.) without any elaboration. Figure 3 presents diagrams of aging threats (HCI, TDDB, etc.) but would be much more effective with a description of what each diagram illustrates. Figure 4 creates confusion by marking the IC Foundry as "Untrusted" but not the subsequent Packaging stage, which should be explained.

3. There are quite a lot of formatting issues: Line 32, 43, 44, 117, 292, 293. Fig 5. Just to name a few.

Other minor comments: The categorization of reliability technologies in Section 3 does not appear to be orthogonal. A clearer taxonomy, perhaps based on the level of application (e.g., hardware-level vs. configuration-level), would be more insightful. This is more of a personal preferences but it might be helpful to think about. The strong statement in Section 2.3 that reconfigurable systems are "often compromised by adversaries" is not supported by a reference, which is required for such an assertion.

The paper has a solid foundational structure but requires an overhaul of its technical content to add depth, a critical analysis of the literature, and significant improvements to its presentation. The authors could focus on revising the paper from a high-level list of topics into a survey that provides deep analysis insight.

Author Response

Reviewer #3, Concern #1: Most of the prior work discussed in this paper is summarized with a simple and superficial summary, rather than a useful analysis. For example, the section on intelligent fault tolerance (lines 428 onwards) mentions a series of papers [85-88] that all use evolutionary methods. The discussion of those works is basically "they use evolutionary method with balancing / tolerance for electromagnetic damages / others". What those concepts are are not delved into. Their specific approaches, effectiveness, and trade-offs are not elaborated or compared. This is insufficient for a high-quality survey.

Author response: Thank you very much for this insightful and constructive comment. We fully agree that a high-quality survey should go beyond listing prior works and instead provide critical analysis, comparative insights, and clear exposition of technical distinctions. The original treatment of the intelligent fault tolerance subsection indeed fell short of this standard.

Author action: To carefully address your concern, we have significantly enhanced the second paragraph of Section 4.2.4 to provide a deeper analytical discussion of the series of works [85–88] that employ evolutionary algorithms for fault tolerance. Rather than offering only high-level descriptions, the revised text now explicitly clarifies the core operational mechanism of each approach, specifies its intended application context and targeted fault model—such as single-event upsets, aging-induced timing faults, or electromagnetic interference—and critically examines its key limitations and quantified trade-offs, including computational overhead, convergence guarantees, area/power penalties, and scalability. This refined analysis ensures that readers gain substantive insight into both the technical distinctions and practical implications of these intelligent fault tolerance strategies, aligning the discussion with the expected depth of a high-quality survey.

 

 

Reviewer #3, Concern #2: The survey is not comprehensive and omits some relevant work. A notable example is the absence of work by Cong and Gururaj on application-level correctness, which represents one perspective on fault tolerance, and the whole group of similar work of soft error handling is not covered.

Author response: Thank you very much for your insightful comment. We acknowledge that the initial version of our survey did not sufficiently cover important contributions in application-level fault tolerance, particularly the work by Cong and Gururaj on ensuring correctness under hardware faults, as well as related research on soft error handling.

Author action: To carefully address your comment, we have expanded Section 4.2 to include a dedicated discussion of application-level correctness approaches, explicitly citing the foundational work by Cong and Gururaj and situating it within the broader context of software- and algorithm-level techniques for soft error resilience. This addition helps present a more balanced and comprehensive view of fault tolerance across the hardware-software stack.

 

 

Reviewer #3, Concern #3: Some concepts lack proper context and citation. The STAR (Self-Test Areas) method is presented as being specific to FPGAs and cites a 2014 paper on bitstream readback. However, the cited paper has no mention of it, and this is in fact a proposal made much earlier at Bell Labs; the original work is not cited, and the context is misleading: it is not a part of existing FPGA devices.

Author response: Thank you very much for your insightful comment. We appreciate the opportunity to correct an important inaccuracy regarding the origin and context of the STAR (Self-Test Areas) method.

Author action: To carefully address your comment, we have revised the discussion of STAR to clarify that it is a research-level fault tolerance architecture originally proposed by Karri, Karpovsky, and Patel at Bell Labs in the late 1990s, not a feature of commercial FPGA devices. We have removed the incorrect citation to the 2014 bitstream readback paper and replaced it with the appropriate original references, ensuring accurate attribution and proper technical context.

 

 

Reviewer #3, Concern #4: Phrases such as "at home and abroad" are not appropriate for international journals, and unconventional terms like "research windfall", "so-called FPGA" should be revised. The overall text would benefit from a proofreading by a native English speaker.

Author response: Thank you very much for this helpful observation. We agree that the phrasing should adhere to the conventions of international academic writing.

Author action: To carefully address your comment, we have removed all non-standard expressions such as “at home and abroad,” “research windfall,” and “so-called FPGA,” replacing them with precise and discipline-appropriate terminology. The entire manuscript has also been thoroughly proofread by a native English-speaking colleague to improve clarity, fluency, and stylistic consistency.

 

 

Reviewer #3, Concern #5: Figure 1 is an art illustration of application areas that adds little value and would be better served by descriptive text with citations. The text mentions military applications, which should also be cited. Figure 2 does not clearly illustrate the Single Event Effects (SEE) it purports to, and the accompanying text lists several SEE types (SEU, SET, SEFI, etc.) without any elaboration. Figure 3 presents diagrams of aging threats (HCI, TDDB, etc.) but would be much more effective with a description of what each diagram illustrates. Figure 4 creates confusion by marking the IC Foundry as "Untrusted" but not the subsequent Packaging stage, which should be explained.

Author response: Thank you very much for your constructive feedback on the figures. We agree that clearer illustration and better integration with the text are needed to enhance their scientific value.

Author action: To carefully address your comment, we have revised Figures 1–4 and their accompanying discussions: Figure 1 has been replaced with descriptive text supported by relevant citations, including for military applications; Figure 2 now includes labeled schematics that clearly differentiate SEU, SET, SEFI, and other SEE types, with concise explanations in the text; Figure 3 has been updated with annotations and captions that explicitly describe what each aging mechanism (HCI, TDDB, etc.) represents; and in Figure 4, we have clarified in both the caption and main text why the IC Foundry is marked as “Untrusted” while acknowledging that packaging may also involve untrusted parties, depending on the threat model.

 

 

Reviewer #3, Concern #6: There are quite a lot of formatting issues: Line 32, 43, 44, 117, 292, 293. Fig 5. Just to name a few.

Author response: Thank you very much for pointing out these formatting inconsistencies. We sincerely appreciate your careful reading of the manuscript.

Author action: To carefully address your comment, we have thoroughly reviewed the entire manuscript and corrected all identified formatting issues, including those on lines 32, 43, 44, 117, 292, 293, and in Figure 5, ensuring consistent typography, spacing, labeling, and adherence to the journal’s style guidelines.

 

 

Reviewer #3, Concern #7: The categorization of reliability technologies in Section 3 does not appear to be orthogonal. A clearer taxonomy, perhaps based on the level of application (e.g., hardware-level vs. configuration-level), would be more insightful. This is more of a personal preferences but it might be helpful to think about. The strong statement in Section 2.3 that reconfigurable systems are "often compromised by adversaries" is not supported by a reference, which is required for such an assertion.

Author response: Thank you very much for your thoughtful suggestion. We agree that a more structured taxonomy would improve clarity, and we appreciate the note regarding the need for supporting evidence.

Author action: In direct response to the taxonomy concern, we have completely reorganized the reliability technology categorization in Section 3​ to establish a truly orthogonal classification framework based on implementation levels.

 

 

Reviewer #3, Concern #8: The paper has a solid foundational structure but requires an overhaul of its technical content to add depth, a critical analysis of the literature, and significant improvements to its presentation. The authors could focus on revising the paper from a high-level list of topics into a survey that provides deep analysis insight.

Author response: Thank you very much for your valuable and constructive feedback. We fully agree that a high-quality survey should offer deep analytical insight rather than a high-level inventory of topics.

Author action: To carefully address your comment, we have undertaken a thorough revision of the technical content throughout the paper, transforming descriptive summaries into critical analyses that compare methodologies, evaluate trade-offs, highlight research gaps, and contextualize advances within the broader field. The presentation has also been refined to ensure clarity, coherence, and scholarly rigor befitting a comprehensive survey.

 

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

This paper is a review article (also known as a survey paper), not a primary research article. Its stated experimental approach is a literature review and an analysis of existing technologies. In fact, the paper provides an overview of reliability in FPGA-based reconfigurable systems. The authors have identified and categorized the three primary challenges in this domain: fault tolerance, aging mitigation, and hardware attack defense. However, while the topic seems quite relevant and the scope is good, the manuscript in its current form appears to be a very preliminary draft and requires major refinements before it can be considered for publication:

1 - The introduction incorrectly describes the paper's own structure. It states the paper concludes with Section 8. The paper actually concludes with Section 9. It completely omits any mention of Section 7, "Reliability evaluation indicators", which is a full section in the paper.

2 - I found that many sections are simply repeating the same information in different ways, so I strongly suggest to avoid duplication and to significantly shorten the paper. In addition  there is some duplicated text from line 1097 to line  1124

3 - incorrect references to rttable for instance the refer to table 2.1 at line 462 but I guess is to table 1 the same apply to several figures references

4 - Many sentences are long and complex, I guess they can be broken to improve the overall readability of the paper. 

Author Response

Reviewer #4, Concern #1: This paper is a review article (also known as a survey paper), not a primary research article. Its stated experimental approach is a literature review and an analysis of existing technologies. In fact, the paper provides an overview of reliability in FPGA-based reconfigurable systems. The authors have identified and categorized the three primary challenges in this domain: fault tolerance, aging mitigation, and hardware attack defense. However, while the topic seems quite relevant and the scope is good, the manuscript in its current form appears to be a very preliminary draft and requires major refinements before it can be considered for publication.

Author response: Thank you very much for your thoughtful assessment. We appreciate your recognition of the paper’s relevance and scope, and we acknowledge that the initial submission required substantial improvement.

Author action: To carefully address your comment, we have comprehensively revised the manuscript to move beyond a preliminary overview, deepening the technical analysis, strengthening the critical discussion of existing works, refining the taxonomy of reliability challenges, and enhancing the overall scholarly rigor to meet the standards expected of a high-quality survey paper.

 

 

Reviewer #4, Concern #2: The introduction incorrectly describes the paper's own structure. It states the paper concludes with Section 8. The paper actually concludes with Section 9. It completely omits any mention of Section 7, "Reliability evaluation indicators", which is a full section in the paper.

Author response: Thank you very much for pointing out this oversight in the introduction. We sincerely appreciate your careful reading of the manuscript.

Author action: To carefully address your comment, we have revised the introduction to accurately reflect the paper’s structure, including a clear mention of Section 7 on reliability evaluation indicators and updating the concluding section reference from Section 8 to Section 9.

 

Reviewer #4, Concern #3: I found that many sections are simply repeating the same information in different ways, so I strongly suggest to avoid duplication and to significantly shorten the paper. In addition there is some duplicated text from line 1097 to line 1124.

Author response: Thank you very much for your careful reading and valuable feedback. We agree that conciseness and avoidance of redundancy are essential for a high-quality survey.

Author action: To carefully address your comment, we have thoroughly streamlined the manuscript by removing repetitive content, consolidating overlapping discussions, and deleting the duplicated passage between lines 1097 and 1124, resulting in a significantly more concise and focused paper.

 

 

Reviewer #4, Concern #4: incorrect references to table for instance the refer to table 2.1 at line 462 but I guess is to table 1 the same apply to several figures references.

Author response: Thank you very much for catching these referencing errors. We appreciate your attention to detail.

Author action: To carefully address your comment, we have carefully checked and corrected all incorrect table and figure references throughout the manuscript, including the erroneous reference to “Table 2.1” at line 462 (now updated to “Table 1”) and similar mismatches elsewhere, ensuring consistent and accurate cross-referencing.

 

Reviewer #4, Concern #5: Many sentences are long and complex, I guess they can be broken to improve the overall readability of the paper.

Author response: Thank you very much for your helpful suggestion. We agree that improving sentence structure enhances readability.

Author action: To carefully address your comment, we have revised the manuscript to break up long and complex sentences into clearer, more concise ones throughout, significantly improving the overall flow and readability of the text.

 

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have followed all my recommendations. The manuscript in its current form can be considered for acceptance in Electronics.

Reviewer 4 Report

Comments and Suggestions for Authors

Overall, the paper has certainly improved, and the authors have responded to my queries. Therefore, I recommend that the paper be accepted.