A Learning Automaton-Based Algorithm for Maximizing the Transfer Data Rate in a Biological Nanonetwork
Dimitrios Makrakis
Round 1
Reviewer 1 Report
The author has developed and presented the idea of using LA to adjust the detection threshold in a CSK modulation and claimed (in the title and in line 7) that it will maximize the data rate. However, the results that are presented do not support this improvement. To be able to observe the improvement of the data rate, the performance of the system should be compared with similar algorithms that are presented in the literature (for example [1] and [2]), in terms of performance indices such as BER, etc. Alternatively, some comparisons can be made to show how the system can outweigh a similar communication scheme which has predefined fixed thresholds. Moreover, it has been mentioned that this protocol will improve the adaptability to environmental changes (line 11). However, the testing scenarios that are modeled, both have fixed channel parameters. Presenting some results for a scenario with time-variant channel characteristics will support the claim of the systems capability to adjust itself with dynamic situations.
To conclude, I do not support publishing this paper unless more performance analysis results that support the claims in the text are added.
References:
[1] A. K. Shrivastava, D. Das and R. Mahapatra, "Adaptive Threshold Detection and ISI Mitigation in Mobile Molecular Communication," 2020 IEEE Wireless Communications and Networking Conference (WCNC), 2020, pp. 1-6.
[2] M. Damrath and P. A. Hoeher, "Low-Complexity Adaptive Threshold Detection for Molecular Communication," in IEEE Transactions on NanoBioscience, vol. 15, no. 3, pp. 200-208, April 2016.
Author Response
We would like to thank you very much for your general positive attitude toward our work entitled “A Learning Automaton-based Algorithm for Maximizing Data Rate in a Biological Nanonetwork” and your encouragement to submit a revised version.
Having seriously addressed, in great detail, all of the comments and recommendations, the initial manuscript has been appropriately revised and significantly improved. For your convenience, all modifications in the resubmitted manuscript are carefully explained and highlighted in yellow color.
In what follows, we exactly detail how we have taken into account the reviewers’ comments and have incorporated them in the revised version of this work.
Issues Raised by Reviewer 1
Point 1: To be able to observe the improvement of the data rate, the performance of the system should be compared with similar algorithms that are presented in the literature (for example [1] and [2]), in terms of performance indices such as BER, etc. Alternatively, some comparisons can be made to show how the system can outweigh a similar communication scheme which has predefined fixed thresholds.
Response 1: The authors appreciate this reviewer’s comment as the additional work highly improved the quality of this manuscript. In the revised manuscript we run simulations comparing the proposed algorithm to the solution proposed in [2]. In addition, we also compared our approach with a fixed threshold solution as proposed and both results are presented in section 4, as section 3 was renamed for better presentation and organization of the manuscript.
Point 2: Presenting some results for a scenario with time-variant channel characteristics will support the claim of the systems capability to adjust itself with dynamic situations.
Response 2: The authors thank the reviewer for this valuable comment. We incorporate the suggestions of the reviewer to our simulation runs so the devised simulation scenarios have different input signaling distribution concerning the message rate injected to the system. In Figure 8 we also changed the number of bacteria per pulse in order to support the claim that the proposed algorithm adjusts itself with dynamic channel configurations
Please feel free to contact us for any additional clarification or information needed.
Best Regards,
The author: Konstantinos F. Kantelis
Reviewer 2 Report
The monography paper presents a Learning Automaton-based algorithm for maximizing data rate in a biological nanonetwork.
The bibliography survey, the theoretic background and simulation deployment are solid.
The author can extend somehow the discussion and conclusions sections in order to more improve the impact of the research work.
Author Response
We would like to thank you very much for your general positive attitude toward our work entitled “A Learning Automaton-based Algorithm for Maximizing Data Rate in a Biological Nanonetwork” and your encouragement to submit a revised version.
Having seriously addressed, in great detail, all of the comments and recommendations, the initial manuscript has been appropriately revised and significantly improved. For your convenience, all modifications in the resubmitted manuscript are carefully explained and highlighted in yellow color.
In what follows, we exactly detail how we have taken into account the reviewers’ comments and have incorporated them in the revised version of this work.
Issues Raised by Reviewer 2
Point 1: The author can extend somehow the discussion and conclusions sections in order to more improve the impact of the research work..
Response 1: The authors would like to thank the reviewer for its valuable comment. In the revised manuscript the both sections were amended and enriched accordingly in order to improve the readership of this work. Based on the additional simulations scenarios that were devised in the updated manuscript, discussion and conclusions sections were updated to include the new findings of this work.
Please feel free to contact us for any additional clarification or information needed.
Best Regards,
The author: Konstantinos F. Kantelis
Reviewer 3 Report
The abstract is poor in describing the main research problem, proposed protocol, simulation scenarios, operational parameters affect on biological nanonetworks, and finally main contributions that come with work.
The introduction must have a paragraph to describe the proposed work of this research as compared with mentioned related works and then gives the main expected contribution that may distinguish this work.
The System Model needs hard work to improve it such as describing deeply the ligand-receptor binding process that was proposed to use here and also need explaining the idea with some figures or diagrams or flowchart as so on. note that this process didnt mention in the abstract while mentioned in the title of this article. Figure 1 needs to describe in details as a basic of this work.
Section 3 supposed the proposed protocol but the I didnt catch the main contribution for the author in this proposed mechanism so I think the author needs to explain and proof the importance of his proposed protocol.
Numerical Simulations and Results must be separete from section 3 and to be section 4 with discussion section, and I think the results are not enough to give us a whole picture about the proposed work importance or verifications. Need to improve this section.
Conclusions need to focus on the main contribution may get from this work as a whole.
Author Response
We would like to thank you very much for your general positive attitude toward our work entitled “A Learning Automaton-based Algorithm for Maximizing Data Rate in a Biological Nanonetwork” and your encouragement to submit a revised version.
Having seriously addressed, in great detail, all of the comments and recommendations, the initial manuscript has been appropriately revised and significantly improved. For your convenience, all modifications in the resubmitted manuscript are carefully explained and highlighted in yellow color.
In what follows, we exactly detail how we have taken into account the reviewers’ comments and have incorporated them in the revised version of this work.
Issues Raised by Reviewer 3
Point 1: The abstract is poor in describing the main research problem, proposed protocol, simulation scenarios, operational parameters affect on biological nanonetworks, and finally main contributions that come with work
Response 1: The author would like to thank the reviewer for the constructive comment. In the revised manuscript the abstract was amended in order to incorporate all the suggestions from the reviewer. So in the revised version all the appropriate information were included in the Abstract in order to improve the readership of this work
Point 2: The introduction must have a paragraph to describe the proposed work of this research as compared with mentioned related works and then gives the main expected contribution that may distinguish this work.
Response 2: The authors would like to thank the reviewer for this valuable comment. The Introduction was updated with two paragraphs in order to provide a description of the proposed algorithm and also put this work in the right context related to the existing work in the literature. For this reason two new works were presented in order to provide a solid ground for comparison in the next sections.
Point 3: The System Model needs hard work to improve it such as describing deeply the ligand-receptor binding process that was proposed to use here and also need explaining the idea with some figures or diagrams or flowchart as so on. Figure 1 needs to describe in details as a basic of this work.
Response 3: The authors appreciate this reviewer’s comment. Taking into consideration all the suggestions from the reviewer, we added all the appropriate information about the ligand receptor binding model and backed it up with an additional figure and new references along with a more analytical presentation in Section 2. We also described in great detail both parts of the Figure 1 and based on the reviewer’s comment for a better description, Figure 1 had been converted to two independent figures in order to provide better and more clear readership.
Point 4: Section 3 supposed the proposed protocol but the I didnt catch the main contribution for the author in this proposed mechanism so I think the author needs to explain and proof the importance of his proposed protocol.
Response 4: The authors would like to thank the reviewer for the constructive comment. In the revised manuscript section 3 was significantly amended and updates with additional information about the main contribution of this work. For both novelties of this work, the algorithm with the hi and lows and the operation of the LA, appropriate additions to the manuscript have been made. These include a new figure and a significant amount of clarifications about the core of this work as pointed out by the new highlighted additions throughout the whole Section 3.
Point 5: Numerical Simulations and Results must be separate from section 3 and to be section 4 with discussion section, and I think the results are not enough to give us a whole picture about the proposed work importance or verifications. Need to improve this section.
Response 5: The authors would like to thank the reviewer for the constructive comment. Taking into consideration the above comment, in the revised manuscript we run simulations comparing the proposed algorithm to an established solution in the existing literature. In addition, we also compared our approach with a fixed threshold solution and both results are presented in section 4 as suggested renamed for better presentation and organization of the manuscript. Furthermore, in this revised version of the proposed work, new simulation scenarios are presented having different input signaling distribution concerning the message rate injected to the system. In Figure 8 we also changed the number of bacteria per pulse in order to support the claim that the proposed algorithm adjusts itself with dynamic channel configurations.
Point 6: Conclusions need to focus on the main contribution may get from this work as a whole.
Response 6: The authors appreciate this reviewer’s comment as the additional work highly improved the quality of this manuscript. In the updated version of the manuscript, the Conclusion section (along with the Discussion) was amended and enriched accordingly in order to improve the readership of this work. Based on the additional simulations scenarios that were devised in the updated manuscript, conclusion section was updated to include the new findings of this work as shown from the yellow highlighted additions.
Please feel free to contact us for any additional clarification or information needed.
Best Regards,
The author: Konstantinos F. Kantelis
Reviewer 4 Report
** the words "data rate" is vague. is it "data rate transfer" or "data rate generation" or "data rate communication"
** Line 16: replace "Referring" by "Referred to"
** Line 51: replace "stated" by "defined" and "on" by "via"
** Line 61:replace "leave" by "exclude"
** Line 62: replace "as a subject of an ongoing study and future extension" by "because it is out of the scope of our work"
** Line 77: replace "Over and above that" by "Moreover"
** Line 78: remove (5)
** Figure 1 (b): replace "stands for" by "shows"
** Figure 1(b) is never discussed in depth or how it was generated
** Figure 1(b) is placed far away from text referencing it
** Line 100: "vide infra" is not English. it is latin, it means "refer reader for later". Remove it as LA is defined in next immediate paragraph. Reader will get it
** Line 103: add "in" after "operate"
** poor English: need a native English speaker to edit paper
** Line: 106: remove "Regarding"
** Line 114: add (b = 0) after (3)
** Figure 2(b): caption is missing a word unless "Interaction" under Figure 2(a) is for Figure 2(b) and is misplaced
** Line 117: replace "from" by "based on"
** Line 143: add ",Figure 2(a)" ** Line after "A"
** Figure 2(A) and (B): add more in-depth analysis for both. What is B signal, what is the Environment", etc.
** Line 154: remove "Contributing to gain a clear picture of the proposed algorithm,"
** Line 157: replace "left" by "Figure 3(a)" and "right" by "Figure 3(b)"
** Figure 3 caption: replace "left" by "(a)" and "right" by "(b)"
** Line 173: replace "again" by this"
** explain why the number of ascensions is linear after the convergence at around 700. And why the stable state was reached at 18,000 seconds
** Line 175: replace "Being a point of segregation from existing works" by "Unlike existing algorithms in the literature"
** Line 180 and 181: explain why the statement is true. What is missing from current work?
** manuscript must be improved both in English grammar and scientific depth before publication
Author Response
We would like to thank you very much for your general positive attitude toward our work entitled “A Learning Automaton-based Algorithm for Maximizing Data Rate in a Biological Nanonetwork” and your encouragement to submit a revised version.
Having seriously addressed, in great detail, all of the comments and recommendations, the initial manuscript has been appropriately revised and significantly improved. For your convenience, all modifications in the resubmitted manuscript are carefully explained and highlighted in yellow color.
In what follows, we exactly detail how we have taken into account the reviewers’ comments and have incorporated them in the revised version of this work.
Issues Raised by Reviewer 4
Point 1: ** the words "data rate" is vague. is it "data rate transfer" or "data rate generation" or "data rate communication"
Response 1: The authors would like to thank the reviewer for this valuable comment. The appropriate modification is the “transfer data rate”. Following the reviewer’s comment all the changes have been made.
Point 2: Line 16: replace "Referring" by "Referred to", Line 51: replace "stated" by "defined" and "on" by "via", Line 61:replace "leave" by "exclude", Line 62: replace "as a subject of an ongoing study and future extension" by "because it is out of the scope of our work", Line 77: replace "Over and above that" by "Moreover", Line 78: remove (5), Figure 1 (b): replace "stands for" by "shows", Figure 1(b) is placed far away from text referencing it, Line 100: "vide infra" is not English. it is latin, it means "refer reader for later", Remove it as LA is defined in next immediate paragraph. Reader will get it, Line 103: add "in" after "operate", Line: 106: remove "Regarding", Line 114: add (b = 0) after (3), Line 117: replace "from" by "based on", Line 143: add ",Figure 2(a)" ** Line after "A", Line 154: remove "Contributing to gain a clear picture of the proposed algorithm," Line 157: replace "left" by "Figure 3(a)" and "right" by "Figure 3(b)", Figure 3 caption: replace "left" by "(a)" and "right" by "(b)", Line 173: replace "again" by this", Line 175: replace "Being a point of segregation from existing works" by "Unlike existing algorithms in the literature"
Response 2: The authors would like to thank the reviewer for bringing these issues to our attention as these changes had enhanced the quality of the manuscript. Based on the reviewer’s suggestions, the revised manuscript has been amended accordingly and rectified ALL the above errors. (Due to the heavily modified version, one or two points may not be part of the new manuscript as the original text has been removed/changed).
Point 3: Figure 1(b) is never discussed in depth or how it was generated
Response 3: The authors would like to thank the reviewer for this comment. Figure 1 has been break to two independent figures and the appropriate additions have been made to provide a clearer description about both the simulation testbed and the different pulse shapes that the algorithm should take into consideration.
Point 4: Figure 2(b): caption is missing a word unless "Interaction" under Figure 2(a) is for Figure 2(b) and is misplaced
Response 4: The authors would like to thank the reviewer for this comment. The caption of the figure was rectified accordingly.
Point 5: Figure 2(A) and (B): add more in-depth analysis for both. What is B signal, what is the Environment", etc.
Response 5: The authors would like to thank the reviewer for the constructive comment. A new set of paragraphs have been added to the revised manuscript providing additional information about the principles of the LA operation (part 2A) and the flowchart of the proposed solution (part 2B). All the additions made were highlighted yellow in the new version of this work.
Point 6: explain why the number of ascensions is linear after the convergence at around 700. And why the stable state was reached at 18,000 seconds
Response 6: The authors would like to thank the reviewer for the constructive comment. Based on the observation of the reviewer, we did a smoothing exercise on the data of the figure 9 and we found that the best fit is from a third degree polynomial function. The Figure was updated accordingly. We believe that the impression about the linear relation is due to absence of additional points.
In figure 9 we present the time needed from the algorithm to find a stable point having as initial value for the number of ascensions the number 2200. Due to the long distance between the two communication parties, each transmission takes more than 250 seconds to complete. Making a rough calculation, this means that the algorithm demanded about 75 transmissions to converge from the 2200 ascensions to the 700. Starting from 1000 ascensions, it took less than 3000 seconds to converge meaning that after 12 messages the algorithm had already found the best number of ascensions for this simulation setup (sweet spot for this simulation configuration). A more aggressive policy could be also applied by changing the number of additional steps to a higher number (from 100 to 300) for every iteration but the higher the value the less accuracy for the sweet spot of the algorithm.
Point 7: Line 180 and 181: explain why the statement is true. What is missing from current work?
Response 7: The authors would like to thank the reviewer for this comment. In order to provide a simulation as close to the “in vitro” environment we simulate the node entity as living cell and this means that the number of receptors that the cell has are finite and in some cases all the receptor could be fully occupied. As a direct result, the node would not sense higher concentration pulses. Adding into the equation the Residence Time (the time a ligand occupies a receptor site) then it becomes clear that end-to-end communication inside a biological communication system is governed by a plethora of parameters. We believe that this study could be the subject of another independent work and we mentioned it to provide some insight about the scientific area of biological nanonetworks.
Please feel free to contact us for any additional clarification or information needed.
Best Regards,
The author: Konstantinos F. Kantelis
Round 2
Reviewer 3 Report
author did a good work to enhance his article ..
