kumoh national institute of technology
Networked Systems Lab.

Rubina, "Low-Latency and Energy-Efficient Trickle-based Scheme for Low-Power & Lossy Industrial WSN"
By : Rubina
Date : 2020-01-14
Views : 190

Dear Prof. Dong-Sung Kim:

The review of the referenced manuscript, WCL2019-1235, is now complete. I regret to inform you that based on the enclosed reviews and my own reading of your manuscript, I am unable to recommend its publication in IEEE Wireless Communications Letters.

Your paper may not be resubmitted for review. The reasons for this are as follows: The reviewer has some major concerns on the novelty of your submission. They are also not satisfied with the literature review since most references are out-of-date. In addition, as pointed out by Reviewer 3, the contribution is rather limited since the proposed approach is optimistic and sub-optimal. The reviewers also suggested that the paper is not well presented with lots of typos grammatical errors. Based on the above-mentioned weakness and with my own reading of your submission, I have to decide to reject your submission.

The reviewers' comments are found at the end of this email.

Thank you for submitting your work to the IEEE Wireless Communications Letters. I hope the outcome of this specific submission will not discourage you from the submission of future manuscripts.

Sincerely,

Prof. Guanding Yu
Editor, IEEE Wireless Communications Letters
yuguanding@zju.edu.cn, yuguanding@gmail.com

Reviewer: 1

Comments to the Author
A technique for solving both the problems(node latency and load-balancing) of Trickle
The algorithm has been presented in this letter.
The scheme of Trickle has been modified to attain better power and energy consumption.
The letter is well written and it is not difficult to understand the contribution.
The proposed solution is technically sound and is demonstrated concretely throughout the paper.
In addition, the performance of the proposed solution is fairly compared with the existing solutions.
Therefore in my opinion, except for some minor points which need to be improved, the letter is almost acceptable.
1) In Algorithm 1(from line 19 to 23) & In Algorithm 2(from line 27 to 31) the author should use only IF statement instead of IF ELSE statement.
As the author explained if inconsistent messages are heard then it simply sets I to Imin and if no inconsistent messages are heard then it
simply doubles the length of the current interval. If the interval
length exceeds the value of Imax, then it directly resets I to Imax only so here there is no need for IF ELSE statement.

2) The authors should point out the eventual shortcomings of the proposed scheme compared to the related existing works.

Reviewer: 2

Comments to the Author
1. This work has made a slight modification upon the standard trickle algorithm. The novelty is quite limited.
2. The authors should clearly present the motivations behind the modifications. For example, random time t is selected from [0, i_min/3^{s+1}] instead of the original trickle range (i/2, i). However. why the upper bound is i_min/3^{s+1}, but not i_min/2^{s+1}?
3. The literature review is so weak. Actually, I can not find any related works. What's more, there is even no reference to these two benchmark algorithms, namely E-trickle and I-trickle.
4. The presentation should be considerably improved. Significant copy editing needs to be done as there are a large number of syntactic and grammatical errors.


Reviewer: 3

Comments to the Author
This paper proposes to modify the trickle timer operation for fast data dissemination in a low-power, lossy wireless sensor network. The modification is a simple choice of reduced period for broadcast timer, t. However the work seems out of time, not enough motivation, not enough substantiation on the proposal and not convincing results. There's no insight to the gains and the article is preliminary in its contribution.

1. The work seems to be outdated as most of the references are 2011. Furthermore, new low-latency protocols such as LWB for WSNs have been proposed. They are not even cited.
2. As stated in the abstract, this proposal is optimistic and sub-optimal. What is the need then for this protocol? Clearly motivate and show the shortcomings of the work
3. Clarify what you mean by IoT is a self-controlled network.
4. There's a sharp difference in writing between the first 2 paragraphs of the rest of the paper. The rest of paper is not well written and with lot of typos.
5. What is the need of tying trickle with RPL? Trickle is a generic timer and can be used for a variety of things.
6. While the title has 'Industrial WSNs', the word is just a random usage and the work does not relate to it.
7. The proposal is to reduce the timer interval within which t is chosen. This has been set to [i/2, i] for a reason as mentioned in the RFC. By changing it, it would obviously reduce latencies but you will sacrifice with a lot of packet losses. This is not even discussed in the paper. Any sort of reasoning is not provided in the paper as to why 3^s is chosen along with any definition for the parameter 's'.
8. While it is clearly mentioned in RFC6206 that changes would bring in sub-optimal results in edge cases, the authors have not cared to evaluate it well. The evaluation is basically a few runs of the simulator without any consideration of whether the simulation is close to reality. The results are not discussed and no insights are provided. The gains indicated are not significant at all.