You are currently viewing a new version of our website. To view the old version click .
MDPIMDPI
Search all articles
by
  • Ionel Petrescu,
  • Elisabeta Niculae and
  • Viorel Vulturescu
  • et al.

Reviewer 1: Anonymous Reviewer 2: Anonymous Reviewer 3: Anonymous

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for assembling so many details into one place. It must have been a very cumbersome task. As a researcher in the IoT field I truly appreciate your work. 

I do think there is one very important aspect which could truly enhance this work: providing tons of examples. For example, it is great that you mention a strength or weakness of a certain protocol. It would be great if you could provide one or more examples of where that protocol is being used and how that said strength is paying off or how the mentioned weakness is providing a hindrance. I know that alone would be hard since you cover so many protocols and that the problem will cascade once you have to provide examples of the stacks of protocols working together. But, the examples would a) provide a real world example or two of that architecture being actually used and b) help your readers find examples that they can further study and maybe even mimic for projects of their own. But also, c) be very interesting if you do not find any examples of certain stacks being used at all, which could result in future work by you or others to figure out why such stacks have not been used or to create examples that do - for further study if nothing else. 

For that reason, I am recommending that the paper be sent back to you for that revision - which is not an easy nor short ask. But the resulting paper you will thus produce will be heavily influential especially for students and newcomers to the field instead of just being a summarized catalog for professionals alone. 

I hope this helps you improve this paper and transform it into an influential text in our field. 

Thanks again and best of luck. 

Comments on the Quality of English Language

Dear Authors,

First, the English of individual sentences and paragraphs is great and informative. However, they do not flow well into each other which makes the readability of the paper very choppy. As a researcher in the field of IoT I can often follow along with the ideas but I wonder how well students or new comers to our field will be able to read and appreciate the amount of work you have put into this paper. 

Second, I think if you used bullet points and itemized lists, you would be able to both format the text more so that it can be visually helpful in separating the discussed concepts, and help solve the aforementioned flow issue. 

Hope these help you revise the writing.

Author Response

Dear Reviewer,


We sincerely thank you for your detailed and very constructive report. Your positive remarks regarding the relevance of the topic and the effort invested in assembling the material are highly appreciated. Your main recommendation—introducing concrete examples and clarifying protocol interactions—has been extremely helpful in strengthening the scientific and educational value of the manuscript.

Following your suggestions, we have substantially revised the paper as follows:

  1. Added numerous real-world examples of protocol deployments, illustrating how specific strengths and weaknesses manifest in practice. These examples appear throughout Section 5 Practical Deployments of IoT Communication Protocols and help readers link conceptual characteristics with concrete use cases.
  2. Expanded the comparative approach by providing examples of protocol stacks and scenarios where protocol combinations succeed or encounter limitations, directly addressing points (a), (b), and (c) of your recommendation.
  3. Enhanced the readability by introducing bullet points, formatted lists, and improved transitions between paragraphs. These changes help create a smoother flow for newcomers to the field, while retaining technical depth for experienced researchers.

Your insight regarding the importance of examples for student and early-career researchers was especially valuable. We believe these additions significantly improve the manuscript’s clarity and impact. We are deeply grateful for your thoughtful guidance, which led to a substantially stronger and hope as you to will apreciate the new form of the paper and accept for published.

 

Kind regards,

Team authors

Reviewer 2 Report

Comments and Suggestions for Authors

The present study conducts a comprehensive review of the main transport and application protocols for the Internet of Things (such as MQTT, CoAP, LwM2M, and others), which define how devices exchange data, maintain reliability, and interact with cloud and enterprise systems. The work examines the design principles, communication models, reliability mechanisms, and security features of each protocol to guide architects and developers in selecting the optimal protocol stack to optimize IoT system performance and its long-term sustainability. The analysis focuses on mapping protocol characteristics with IoT requirements, including scalability, interoperability, and energy efficiency. It concludes that, although MQTT dominates cloud telemetry, CoAP and LwM2M are superior in IP-based constrained networks, and protocols such as OSCORE are central for end-to-end security. 
However, considering that the article focuses on the most widely implemented IoT communication protocols and relies on well-established specifications and the synthesis of already published empirical evaluations on energy efficiency, the question arises as to whether the present work truly adds something new to the literature, or if it primarily serves as a descriptive reaffirmation and a synthesis of already known trends and trade-offs in the area. Avenues for transforming the work into a more robust production, in my understanding, would include addressing the following issues:

1. The original title sounds inadequate and even incomplete to me, as it omits the article's focus also on the Transport layer. By focusing on the Application layer, where the main protocols (MQTT, CoAP, LwM2M, OPC UA) reside, the title diverges from the article's text, which demonstrates that the scope of the review is broader.

Recommendation to authors: Modify the title to a new one that more accurately reflects the content and the declared scope of the article.

2. Table 1 does not present any novelty to the literature on the subject. Although it summarizes the trade-offs of each protocol in terms of range, throughput, power consumption, and topology, it does so in a qualitative and generic way (some classifications are even questionable). The primary synthesis of the work, therefore, does not add a new taxonomic model or a quantifiable meta-analysis that justifies publication as an original contribution to knowledge.

Recommendation to authors: Substitute or complement Table 1 with a quantitative meta-analysis table. This table should synthesize empirical data (instead of qualitative labels) from studies cited in Section 4, comparing crucial metrics such as: a) compare MQTT (QoS 0, 1, 2), CoAP (Confirmable vs. Non-Confirmable), and HTTP/HTTPS across different device classes (microcontrollers vs. more capable edge devices), as already suggested by the cited studies; b) compare the handshake time (TLS/DTLS) and message latency; c) quantify the overhead in bytes (e.g., CoAP 4 bytes vs. verbose HTTP); d) demand that the energy efficiency hierarchy (CoAP/MQTT-SN are superior to MQTT/HTTP) be presented graphically, using data extracted from the cited research articles.

3) Deepening the Research Agenda - Section 5 already points to important future directions, but these must be connected to the analysis of current protocol limitations.

Recommendation to Authors: The article mentions PQC as a future direction. The authors should discuss how the resource limitations of current protocols (the overhead of TLS/DTLS) would be exacerbated by PQC algorithms, which tend to require greater computational power and key sizes, thereby intensifying the need for lightweight solutions like OSCORE. The article must go beyond merely mentioning integration with 6G and URLLC (Ultra-Reliable Low-Latency Communication). An explicit analysis must be conducted on why existing protocols (MQTT, CoAP) — which face challenges related to reliability, interoperability, and QoS — are not natively suitable for deterministic and mission-critical requirements (e.g., precision industrial automation), which justifies the search for new deterministic networking mechanisms. Expand the idea of AI-driven adaptive protocol selection mechanisms. This must be framed as a direct response to the fact that there is no single universal solution (one-size-fits-all).

Comments on the Quality of English Language

English should be improved. Issues requiring review:

1. Grammatical inconsistencies. Example: the phrase "Currently, there is not universally reference architecture for the IoT" needs correction for proper grammar.

2. Unnatural phrasing/non-idiomatic construction: The wording "guides on selecting the optimal protocol stack" could be rephrased for better flow.

Author Response

Dear Reviewer,
We greatly appreciate your thorough assessment of our manuscript. Your comments provided essential direction for deepening the analysis, refining the structure, and ensuring the article offers a meaningful contribution beyond existing literature.

We have fully addressed your recommendations:

  1. Title revised.
    In line with your observation that the review covers both Transport and Application layer protocols, we have changed the title to:
    “Transport and application layer protocols for IoT: A comprehensive review.”
  2. Significant enhancement of comparative analysis.
    To enrich the original Table 1 and respond to the need for quantitative synthesis, we added:
  • New comparative tables (Tables 2 and 3) including empirical measurements derived from cited studies, covering QoS variants, confirmable vs non-confirmable messages, RTT, handshake latency, and protocol overhead.
  • New figures (Figures 2 and 3) illustrating energy consumption patterns, payload overhead, and latency behavior.

These additions transform the comparative section into a structured and data-driven analysis, addressing your point that the manuscript must contribute beyond descriptive summaries.

  1. Expanded discussion on future research directions.
    Section 6 Conclusions now includes explicit treatment of:
  • The implications of post-quantum cryptography (PQC) on constrained IoT devices and security protocols such as DTLS/TLS.
  • The mismatch between current protocols and the deterministic requirements of URLLC, industrial automation, and 6G systems, explaining why new paradigms are needed.
  • The emerging potential of AI-driven adaptive protocol selection mechanisms as a response to the lack of universal solutions.
  1. Language improvements.
    All grammatical issues highlighted (e.g., “there is not universally reference architecture”) were corrected, and the phrasing throughout the paper was refined for clarity and idiomatic correctness.

Your constructive feedback guided major improvements in analytical depth and methodological rigor. We sincerely thank you for helping us elevate the manuscript to a higher scientific standard, and hope as you to will apreciate the new form of the paper and accept for published.

 

Kind regards,

Team authors

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript presents a wide-ranging review of application and transport protocols used in IoT systems. The topic is timely and the set of protocols selected is indeed representative of the current technological landscape. The paper demonstrates substantial effort in gathering information from technical documentation, standards, and recent scientific publications. Despite these strengths, several areas require improvement to increase the paper’s scientific rigor, clarity, and overall coherence.

A first issue concerns the structure of the review. Although each protocol is discussed in its own subsection, the depth and orientation of the analysis vary considerably across sections. Some protocols are described mainly through functional characteristics, while others include deployment considerations, security implications, or energy profiles. As a result, it becomes difficult for readers to compare protocols in a consistent manner. The manuscript would benefit significantly from a unified analytical framework applied to all protocols—e.g., a fixed set of dimensions such as transport characteristics, message model, reliability, security mechanisms, device constraints, and typical deployment scope. Applying the same evaluative grid throughout would facilitate meaningful comparison and strengthen the scientific quality of the review.

The methodological approach used for selecting the literature and protocols is not clearly explained. This reduces transparency and makes it difficult to assess the completeness of the review. The introduction should include a short but explicit explanation of how the authors identified and selected sources (protocol specifications, surveys, empirical studies) and why the chosen set of protocols is considered representative. Without this clarification, the review risks being perceived as a compilation of protocol descriptions rather than a systematic or semi-systematic scholarly study.

Several sections contain technical descriptions that closely resemble those commonly found in online documentation or introductory material. While the information is correct, the academic value of the review would be enhanced by deeper critical interpretation. For instance, the paper could more explicitly discuss issues such as protocol scalability under real-world constraints, the concrete impact of security options on device lifetime, or the conditions under which a protocol’s theoretical advantages become negligible in practice. Engaging more directly with the empirical studies cited—rather than merely listing their conclusions—would also add depth and originality.

The manuscript includes a comparative table, which is a useful addition, but the criteria appear to mix heterogeneous categories (e.g., “latency,” “security,” “power consumption” are all important but measured differently). The table would be clearer if the comparison criteria were better justified or if the authors explained how qualitative judgments (low, moderate, high) were assigned. Otherwise, the reader may find it difficult to understand the basis for the evaluations.

In various parts of the manuscript, the writing style suffers from repetition, long sentences, and inconsistent terminology. For example, the introductions to sections sometimes restate concepts already explained earlier in the paper, which affects readability. The discussion and conclusion should also be rewritten to emphasize the insights produced by the review rather than repeating general statements about IoT challenges. Strengthening the conclusion with a concise summary of protocol suitability across distinct application domains would enhance the practical relevance of the work.

Lastly, while the paper cites a substantial number of recent references, the narrative does not always integrate them effectively. The authors should clarify how each cited work contributes to their analysis—for instance, whether a reference is used to support a performance claim, illustrate a use case, or provide empirical evidence. Strengthening the link between the literature and the authors’ comparative assessment would give the manuscript a more solid scholarly foundation.

Finally, the paper addresses an important subject and assembles a useful amount of information. To reach the publication standards of the journal, it is recommended that the authors (1) adopt a more systematic comparison framework, (2) clarify their methodology for selecting protocols and sources, (3) increase the analytical depth of the discussion, (4) revise the language for precision and conciseness, and (5) articulate more clearly what new insight this review offers compared to previous surveys. With these improvements, the manuscript could become a valuable contribution to the IoT protocol literature.

Author Response

Dear Reviewer,


We are truly grateful for your comprehensive and insightful review. Your comments addressed core aspects of academic rigor, consistency, and methodological transparency, and they significantly helped us improve both the structure and scholarly value of the manuscript.

We have addressed each of your recommendations in detail:

  1. Unified analytical framework.
    To ensure comparability across protocols, we revised Sections 4 and 5 by applying a common evaluative structure to each protocol:
  • transport characteristics
  • message model
  • reliability mechanisms
  • security features
  • device constraints
  • typical deployment environments

This consistent grid allows readers to perform meaningful cross-protocol analysis and enhances the scientific coherence of the review.

  1. Clarification of methodology.
    The introduction now describes the systematic approach used to select protocols and literature sources, explaining the inclusion criteria for specifications, surveys, and empirical studies.
  2. Increased analytical depth.
    The paper now explicitly discusses:
  • protocol scalability in real-world deployments
  • empirical insights on energy impact
  • trade-offs between theoretical and practical performance
  • conditions under which certain advantages become negligible

These additions transform descriptive sections into critical analysis supported by recent literature.

  1. Improved comparative components.
    We expanded the comparative section with new tables and figures (Tables 2, 3; Figures 2, 3), clearly motivated and explained in the text. Qualitative assessments are now supported by quantitative data and referenced empirical results.
  2. Language refinement.
    Repetitions, long sentences, and inconsistent terminology have been addressed. The discussion and conclusions were rewritten to emphasize insights and practical implications.
  3. Clear articulation of contributions.
    The paper now outlines its contribution relative to prior surveys, highlighting the integration of quantitative comparisons, deployment-oriented examples, and a unified analytic framework.

We are sincerely thankful for your thoughtful review. Your observations prompted substantial enhancements that greatly improved the clarity, rigor, and impact of the paper, and hope as you to will apreciate the new form of the paper and accept for published.

 

Kind regards,

Team authors

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

As suggested, authors revised the title (to "Transport and Application Layer Protocols for IoT: A Comprehensive Review") to better reflect the review's scope across both Transport and Application layers. More significantly, they introduced a significant enhancement to the comparative analysis. This involved adding new comparative tables (Tables 2 and 3) that incorporate empirical measurements from cited studies, covering metrics such as QoS variants, confirmable versus non-confirmable messaging, RTT, handshake latency, and protocol overhead. Furthermore, they added new figures (Figures 2 and 3) to illustrate patterns in power consumption, payload overhead, and latency behavior, transforming the comparative section into a structured, data-driven analysis that moves beyond mere descriptive summaries, and they also included an expanded discussion on future research directions.

The article's conclusion improved significantly with the inclusions, transforming it into a detailed prospective analysis rather than a mere summary of the findings. The improvements added technical depth, such as the discussion of the eeDTLS protocol, which was suggested for reducing UDP/IP and DTLS header overhead by 91% (from 77 bytes to 7 bytes), consequently decreasing the energy consumption associated with DTLS handshake computations. Additionally, the conclusion now details critical future research directions, including the necessity of improving interoperability and integrating protocols with emerging 6G networks (URLLC), the challenge of implementing robust security against quantum threats using Post-Quantum Cryptography (PQC), which necessitates the use of lightweight approaches like OSCORE, and the development of AI-driven adaptive protocol selection mechanisms to optimize communication efficiency.

The new version of the article also incorporated language improvements to increase clarity, conciseness, and academic style, eliminating verbose and redundant constructions.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors answered to the reviewer's comments. The paper can be published in the current form.