Review Reports

Round 1

Reviewer 1 Report

It is very hard to summarize what the paper is proposing, since (i) it is very hard to read, and (ii) the motivation are not explained at all. Due to its weaknesses, I would recommend a deep revision of this paper, which, in its current for, is not ready for publication.

==== On the writing ====
Overall the paper is extremely hard to read: it contains many sentences that are simply not correct in plain English. In many parts of the paper, it is extremely difficult to understand what the authors are saying. Since it is not possible to understand what the authors are presenting, it is not possible to assess correctly many parts of the paper. The paper requires an overall revision with the help of a English native speaker.

==== On the motivations ====
The paper fails at presenting in a clear and well organized way the problem the authors are trying to solve, along with the motivation on the relevance of such a problem. In addition they do not discuss properly the related work.
Why reaching an agreement in a IoT scenario is important? Which types of applications need such an agreement? Why should we need a distributed approach and which scenario the authors are targeting? What are the limitations of the previous solutions and why can not they be used in this scenario? Scientific papers should follow a widely recognized convention where the problem that the paper is trying to solve is clearly defined, where the related work are discused in depth, where the solution is clearly presented, and where such a solution is evaluated. The authors did not even bother to test their solution on practical settings (e.g., using a simulator).

==== Summary ====
In summary, the paper seems to be a collection of confused ideas, without a proper scientific approach. I strongly suggest to revise and reorganize the paper before any future submission.

Author Response

Response to Reviewer 1 Comments

 

Point 1: It is very hard to summarize what the paper is proposing, since (i) it is very hard to read, and (ii) the motivation are not explained at all. Due to its weaknesses, I would recommend a deep revision of this paper, which, in its current for, is not ready for publication.

Response 1:

Thank you for your kindness to read our paper. The motivation of the paper is try to propose a brand new protocol to make all normal PUs reach an agreement underlying the MEC-based CIoT platform, where MEC means mobile edge computing and CIoT is abbreviated from cellular internet of things. With the agreement, many applications can be implemented, such as the task of locating the location of the copied files in a distributed environment [12,13], a two-phase commitment can be made in the distributed database system [14,15] and the landing task controlled by the flight control system [16,17] The agreement problem has been also studied and applied extensively in diverse areas including control systems, optimization, distributed computing, and emerging areas such as robotics, Block chain, and IoT [18,19]. ▊

 

 

Point 2: Overall the paper is extremely hard to read: it contains many sentences that are simply not correct in plain English. In many parts of the paper, it is extremely difficult to understand what the authors are saying. Since it is not possible to understand what the authors are presenting, it is not possible to assess correctly many parts of the paper. The paper requires an overall revision with the help of a English native speaker.

Response 2:

The revision has been checked with the help of an English native speaker. ▊

 

 

Point 3: The paper fails at presenting in a clear and well organized way the problem the authors are trying to solve, along with the motivation on the relevance of such a problem. In addition they do not discuss properly the related work.
Why reaching an agreement in a IoT scenario is important? Which types of applications need such an agreement? Why should we need a distributed approach and which scenario the authors are targeting? What are the limitations of the previous solutions and why can not they be used in this scenario? Scientific papers should follow a widely recognized convention where the problem that the paper is trying to solve is clearly defined, where the related work are discused in depth, where the solution is clearly presented, and where such a solution is evaluated. The authors did not even bother to test their solution on practical settings (e.g., using a simulator).

Response 3:

To reach an agreement is important because all processors in a distributed system can then execute their own task with the same initial value, such as the time stamp, initial state, initial location, and initial database. All existed protocol can solve the agreement problem only underlying their typical network topology. Based on different network topology and different elements of the network structure, all previous protocols cannot be suitable for the MEC-based CIoT platform. The agreement problem has been extensively studied over the past two decades. Many graceful agreement protocols have been proposed with various network topologies [10,11,20,22-28]. In [28], the network topology is Broadcasting Network (BCN); in [11,23], the network topology is Fully Connected Network (FCN); in [10], the network topology is Cloud Computing environment (CC); in [20], the network topology is the Integrated Fog Cloud IoT (IFCIoT). However, the existence of abnormal PUs in the network topology is assumed by these studies [10,11,20,22-28]. Table 2 shows a comparison of these protocols. Among them, only the CIoTAP proposed in this research can be used in a cellular network that implements mobile edge computing and IoT, so that all normal PUs can reach the agreement and provide a highly reliable CIoT platform. To improve the readability of the proposed protocol, an example is given in Section 4. Finally, the complexity and the optimality of the protocol had been evaluated in Section 5.

Because the agreement problem is a theoretical problem, many related studies have proved the optimality of agreement problem through mathematics in the past without any experiments or simulation [10,11,20,22-28]. Therefore, in our research, the optimality of CIoTAP has been demonstrated. The complexity of CIoTAP is proven based on two points, including: 1) The times of data exchange required by CIoTAP is the minimum, and 2) the maximum number of faulty PUs that can be allowed underlying CIoTAP to reach an agreement. Through the proof of these two requirements, CIoTAP is proven to be the optimal solution to the agreement problem under the MECIoT framework. ▊

 

Point 4: In summary, the paper seems to be a collection of confused ideas, without a proper scientific approach. I strongly suggest to revise and reorganize the paper before any future submission.

Response 4:

Thank you for your considerable comments, the paper had been revised and reorganized, and checked with the help of an English native speaker. ▊

Author Response File: Author Response.pdf

Reviewer 2 Report

The article Enhancing the Reliability of Cellular Internet of Things through Agreement is focused on designing and evaluation of the CIoTAP protocol, which is based on the agreement principle and is designed for mobile edge computing IoT applications (MECIoT).

The first part of the article contains an introduction into the topic and evaluation of existing methods, techniques and the description of the environment. The rest of the paper deals with designing and presentation of the CIoTAP protocol and its evaluation.

Generally, the language throughout the entire paper is acceptable and the article is technically sound.

I agree that the introduction into this topic and description of the MECIoT environment is necessary, however, I would recommend to shorten this part of the article a bit. On the other hand, the contributions of the authors, especially the principle of their CIoTAP protocol could be better described and this part of the article should be extended.

For example, I miss any simulation or emulation of their protocol. Its network performance, practical implementation questions, the impact of their protocol on network characteristics (additional delay, overheads, reliability of the network, etc.). The comparison in Tab. 2 is also quite brief and I recommend to provide better comparison with existing techniques and their limitations.

Author Response

Response to Reviewer 2 Comments

 

Point 1: The article enhancing the Reliability of Cellular Internet of Things through Agreement is focused on designing and evaluation of the CIoTAP protocol, which is based on the agreement principle and is designed for mobile edge computing IoT applications (MECIoT).

Response 1:

Thank you for your affirmation of our research topics and results. ▊

 

 

Point 2: The first part of the article contains an introduction into the topic and evaluation of existing methods, techniques and the description of the environment. The rest of the paper deals with designing and presentation of the CIoTAP protocol and its evaluation.

Response 2:

Thank you for the research structure of this paper for your consent. ▊

 

Point 3: Generally, the language throughout the entire paper is acceptable and the article is technically sound.

Response 3:

Thank you for your affirmation. ▊

 

 

Point 4: I agree that the introduction into this topic and description of the MECIoT environment is necessary; however, I would recommend to shorten this part of the article a bit. On the other hand, the contributions of the authors, especially the principle of their CIoTAP protocol could be better described and this part of the article should be extended.

Response 4:

(1) The introduction of the MECIoT environment has been shortened to improve the readability.

(2) The principle of CIoTAP has been described and extended in the revision. ▊

 

 

Point 5: For example, I miss any simulation or emulation of their protocol. Its network performance, practical implementation questions, the impact of their protocol on network characteristics (additional delay, overheads, reliability of the network, etc.).

Response 5:

Because the agreement problem is a theoretical problem, many related studies have proved the optimality of agreement problem through mathematics in the past without any experiments [10,11,20,22-28]. Therefore, in our research, the optimality of CIoTAP has been demonstrated. The complexity of CIoTAP is proven based on two points, including: 1) The times of data exchange required by CIoTAP is the minimum, and 2) the maximum number of faulty PUs that can be allowed underlying CIoTAP to reach an agreement. Through the proof of these two requirements, CIoTAP is proven to be the optimal solution to the agreement problem under the MECIoT framework. ▊

 

 

Point 6: The comparison in Tab. 2 is also quite brief and I recommend to provide better comparison with existing techniques and their limitations.

Response 6:

The agreement problem has been extensively studied over the past two decades. Many graceful agreement protocols have been proposed with various network topologies [10,11,20,22-28]. In [28], the network topology is Broadcasting Network (BCN); in [11,23], the network topology is Fully Connected Network (FCN); in [10], the network topology is Cloud Computing environment (CC); in [20], the network topology is the Integrated Fog Cloud IoT (IFCIoT). However, the existence of abnormal PUs in the network topology is assumed by these studies [10,11,20,22-28]. Table 2 shows a comparison of these protocols. Among them, only the CIoTAP proposed in this research can be used in a cellular network that implements mobile edge computing and IoT, so that all normal PUs can reach the agreement and provide a highly reliable CIoT platform. ▊

 

Reviewer 3 Report

The authors present a good study of CIoT. The paper requires minor changes before it can be accepted. 

1) The authors stated, "In some time-sensitive applications, connecting to the Internet will not be able to meet the time budget, so command and control will reside on the cellular BS in the form of MEC PU." Could this be expanded to better understand the intended message? Also, for this part "Cellular networks provide ubiquitous connectivity, which provides an extremely low probability of outages that traditional wireless networks cannot achieve." Please provide a reference or remove this statement. 

2) The authors claimed that the characteristics of CIOT PU is low power consumption. Could you explain more about this?

3) It will be best to represent using a flow chart the details of CIoTAP as this will clearly emphasize key outcomes.

4) What could be the potential failure points or in other words limitations of the proposed algorithm?

5) How challenging will it be to use the proposed algorithm to replace existing algorithms for a potential IoT application?

6) The 3 assumptions for which this algorithm works for a connected topology is most likely an ideal case. Will the algorithm completely fail if one of the assumptions is violated or can it have a tolerance?

Others:

- Please maintain the font size for all the captions and texts in the figures. 

Author Response

Response to Reviewer 3 Comments

 

The authors present a good study of CIoT. The paper requires minor changes before it can be accepted.

 

Point 1: The authors stated, "In some time-sensitive applications, connecting to the Internet will not be able to meet the time budget, so command and control will reside on the cellular BS in the form of MEC PU." Could this be expanded to better understand the intended message? Also, for this part "Cellular networks provide ubiquitous connectivity, which provides an extremely low probability of outages that traditional wireless networks cannot achieve." Please provide a reference or remove this statement.

Response 1: Based on another referee’s comments, the paragraph of these sentences is removed to improve the readability of the section. ▊

 

Point 2: The authors claimed that the characteristics of CIoT PU is low power consumption. Could you explain more about this?

Response 2: Based on the research results of Qi et al. [21] and US Patent No. 9,693 [*], the CIoT devices may include devices with limited coverage and/or processing, such as devices that mainly run MTC or M2M (e.g., sensor devices, controller devices, etc.) can be known. These CIoT devices can have limited or no user interface, and can be used for machines or devices with little mobility. Therefore, CIoT devices can be deployed in usage scenarios such as home automation (e.g., security, appliances, energy packages, etc.), industry automation, and smart cities with low-power devices (e.g., devices having a battery life of several years), and can be easily installed and operated in challenging coverage conditions, such as lower or basement levels of buildings. ▊  

[21] Qi, Q.; Chen, X.; Zhong, C.; Zhang, Z. Physical tier security for massive access in cellular Internet of Things. Sci. China Inf. Sci. 2020, 63(2), 121301.

[*] Adrangi, F.; Venkatachalam, M.; Jain, P. Procedures to provision and attach a cellular internet of things device to a cloud service provider. U.S. Patent No. 9,693,178. 27, Jun. 2017, Washington, DC: U.S. Patent and Trademark Office.

 

Point 3:  It will be best to represent using a flow chart the details of CIoTAP as this will clearly emphasize key outcomes.

Response 3: Thank you for your comments. Due to the complexity and the large number of text descriptions, the proposed protocol will be difficult to display them on the same page if the flow chart is used, and the text fonts are also bound to be small and difficult to read. In addition, all past researches on agreement issues [10,11,20,22-28] are also use the format as the study presented. ▊

 

Point 4: What could be the potential failure points or in other words limitations of the proposed algorithm?

Response 4: The limitations of the proposed protocol have been explained in the Section 6 conclusion. Underlying different network topology, the proposed protocol may not be suitable. And the proposed protocol may not be used to the environment of both PUs and transmission media are fallible simultaneously. All of these limitations are the future studies. ▊

 

Point 5: How challenging will it be to use the proposed algorithm to replace existing algorithms for a potential IoT application?

Response 5: If the proposed protocol is used to replace existing protocol for potential IoT application, the time complexity will be changed based on the topology of the IoT application itself. And the fault tolerant capability will be lower down depending on the symptom of the processing elements of the IoT application. The best way is to propose a new protocol for the new network topology. ▊

 

Point 6: The 3 assumptions for which this algorithm works for a connected topology is most likely an ideal case. Will the algorithm completely fail if one of the assumptions is violated or can it have a tolerance?

Response 6: It is exactly as referee said the three assumptions for the protocol worked are the limitation of all studies of reaching agreement problem. The assumption 1 & 3: If the PU cannot be identified and not unique, then the receiving PU cannot identify the sending PU for data transmission. The influence of abnormal PU may not be removed, and then the agreement cannot be reached. So assumptions 1 and 3 must be satisfied and are the limitation of the most studies of the agreement problem.

The assumption 2: In the research of Fisher & Lynch [11], it has been proved that "in a distributed computing system with n PUs (n≥4), at most one third of the PUs can fail" must be satisfied. In other words, the agreement cannot be reached as if the number of failed PUs is more than one third of all PUs. Therefore, in order to solve the agreement problem in ECIoT, this assumption must be met; otherwise the agreement problem cannot be solved in ECIoT. ▊

 

Point 7: Please maintain the font size for all the captions and texts in the figures.

Response 7: Thank you for your kindness mention. The font size of the title and text in the figures has been adjusted. ▊

 

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The authors sufficiently addressed my comments and questions and the revision of their paper is acceptable.

Author Response

Response to Reviewer 2 Comments

 

Point 1: The authors sufficiently addressed my comments and questions and the revision of their paper is acceptable.

Response 1: Thank you for your affirmation of our research topics and results.

 

Point 2: English language and style are fine/minor spell check required.

Response 2: A person whose native language is English had reviewed the grammar and the whole paper carefully.

 

Point 3: Are the results clearly presented?  Can be improved.

Response 3: The results had been modified again to make them be clearly presented.

 

Point 4: Are the conclusions supported by the results?  Can be improved

Response 4: The section of conclusion had been reorganized to improve the readability of the results.

 

Author Response File: Author Response.docx