kumoh national institute of technology
Networked Systems Lab.

Energy Efficient and Throughput-aware Relay selection Scheme for Industrial WSN
By : Damian
Date : 2019-10-29
Views : 302

IEEE Transactions on Industrial Informatics





Decision Letter (TII-19-4323)

From:

renluo@ntu.edu.tw

To:

eic.tii@ira.ee.ntu.edu.tw

CC:

Subject:

IEEE Transactions on Industrial Informatics - Manuscript No. TII-19-4323

Body:

Dear Authors (CC to Associate Editor, Reviewers),

Thank you very much for submitting manuscript No. TII-19-4323: "Energy Efficient and Throughput-aware Relay selection Scheme for Industrial WSN" to the IEEE Transactions on Industrial Informatics as a Regular Paper submission.

The review process of your manuscript referenced above has been completed. Much to my regret, I have to inform you that in the opinion of the reviewers and Associate Editor in charge, the submitted manuscript is not suitable for publication in the IEEE Transactions on Industrial Informatics.

For your reference, these comments of the reviewers are enclosed.

I look forward for your contribution, and hope to work with you again.

With my best wishes for your success,


Sincerely yours,

Dr. Ren Luo
Editor-in-Chief
IEEE Transactions on Industrial Informatics
http://tii.ieee-ies.org/
renluo@ntu.edu.tw

****************************************************
Encl.:
Reviewer: 1

Comments to the Author
The work presents an approach for efficient relay selection in IWSNs. The selection process is based on a routing metric function that considers the residual energy and link reliability for selection the best relay node. The results show improvements in terms of energy consumption, throughput and delay, however the reviewer has the following comments:

- Since the paper includes many mathematical symbols, it would better if the authors could include a list of all those symbols.

- Would be better to have a separate section for the related work included in the introduction part.

- Although the authors include a comprehensive review of the related work, they should clearly emphasize how the proposed work is different of what has been presented in the literature.

- WirelessHART and ISA100.11a have been introduced as reliable and energy-efficient standards for IWSNs. The authors need to further highlight why routing protocol adopted in these standards are not applicable for the considered time-sensitive IWSNs?

- Is the energy consumed by transmitting the HELLO messages and their ACKs are considered in the presented energy model?, if not, to how extent these control messages add to the energy consumption.

- It is not clear whether the denominator of eq. (12) represents the number of hello messages, the number of ACKS or the sum of both?!

- It is possible that Hello messages and/or the corresponding ACKs are not received. Does these packet loss has a significant effect on the proposed approach? Why?

- As mentioned in the paper, the rely selection process is for every transmitted packet. How would that affect the energy consumption for industrial applications that have high refresh rates.

- It can happen that many nodes attach to the same rely node, due to its low hop-count, good link quality and high residual energy, which in turn causes a congestion to that node and adversely affects the network performance. How the proposed approach could deal with such problem?

- It would be better if the authors introduce a brief complexity analysis of the proposed work including the added computational complexity, storage and communication overhead. That is to highlight the trade-off for the acquired performance improvements.

- Is the 35m communication range considered in the simulation is practical value for the short range UWB communications?

- What is the confidence interval of the obtained results?

- How the End-to-End delay values are calculated in the simulations, average or worst-case?

- Overall revision of the paper writing and structure is needed.


Reviewer: 2

Comments to the Author
The authors present a routing strategy which takes into acount the energy consumption required to relay the data.
The authors claim this is of use for industrial wsn but this is a fairly weak claim as they do not connect any of their results to industrial application requirements. The mention time sensitive applications as a target application for the proposed solution but without connecting to what is required. The same is true for reliability (the packet delivery ration in the paper).
This is also furthermore being questioned since there are no references to work done for IWSN by industrial researchers at, e.g. Emerson, ABB, etc.

There are several unclear parts in the paper:
- Does this approach take into account the whole path before selecting the next hop or not?
- What is the overhead in terms of energy to compute the paths: there are several non data messages sent to estimate the best path: the Hello(w) message, the RReq, RAck and information messages. Why not combine a few of these messages?
- It is unclear what the authors mean with end-to-end delay? The packet overhead to determine the path must have a sever impact on this? In industial applications, end-to-end delay is normally measured from when the sensor data is being generated at the sensor in the source node until the data can be used at the destination, not from when it is being transmitted (after determining the optimal path) to when it is being received.
- Is Algorithm 1 being performed in all nodes along the route before the data packet is being transmitted from the source?
- It is unclear if the experiments use the IEEE 802.15.4 radio or just the CSMA/CA algorithm (which is not exclusive for 802.15.4)?
- Looking at table 1: simulation parameters it cannot be the 802.15.4 radio used for simulation as several parameters (data packet size, data rate) does not correspond to the 15.4 specs. Which radio is then simulated?
- For the experiments it is unclear what a round is? The authors state that it is "time allocated to transmit data packets from source to destination". What does this mean? Allocated or time taken?
- The experimental results in Fig. 6 is unexplained: why does the end-to-end delay increase for over the number of rounds? The same is true for Fig.8: why does the energy consumption increase? I guess i dont understand what a round is.


Reviewer: 3

Comments to the Author
In this paper, the authors proposed a routing metric to enhance the energy efficiency and throughput of IWSNs. Several major issues are found in this work:
1) The authors claim that the proposed method is for industrial WSN. However, I did not see anything particular for industrial settings or environments. It seems for general WSN, not for IWSN.

2) The proposed scheme required positioning information by using "Ad hoc positioning system". Firstly, it requires additional energy. Secondly, I doubt the accuracy of the positioning information by using that localization scheme. Moreover, the authors are using 802.15.4a. One of the main advantages of UWB is localization. So why not have your own localization scheme?

3) The network model seems an outdoor model. However, most of the industrial environments are indoor, full of various obstacles. Moreover, the authors are using 802.15.4a. Due to the weak penetration ability of UWB, I do not think this model works in industrial environments.

4) No implementation. UWB chips are available now. Please consider real implementation and measurements in industrial environments.


Reviewer: 4

Comments to the Author
For my comments pls. refer to the attached file.


AE Comments: Reject
Associate Editor
Comments to the Author

Date Sent:

27-Oct-2019

File 1:

- review.pdf

Files attached

review.pdf




© Clarivate Analytics | © ScholarOne, Inc., 2019. All Rights Reserved.