kumoh national institute of technology
Networked Systems Lab.

Edge Computational Task Offloading Scheme using Reinforcement Learning for IIoT Scenario
By : Sajjad
Date : 2020-05-11
Views : 306

Dear Mr Hossain,
You have been listed as a co-author of the following submission:
Submission no: ICTE_2020_165
Submission title: Edge Computational Task Offloading Scheme using Reinforcement Learning for IIoT Scenario
Corresponding author: Professor Dong-Seong Kim

We are writing to let you know the status of this submission has changed to Revision Requested. The link below takes you to a webpage where you can log in to our submission system using your existing Elsevier profile credentials or register to create a new profile. You will then have the opportunity to view the submission status and see reviewer and editor comments once they become available.


Once again, thank you very much for your submission.
ICT Express

Ref: ICTE_2020_165
Title: Edge Computational Task Offloading Scheme using Reinforcement Learning for IIoT Scenario
Journal: ICT Express
Dear Professor. Kim,
Thank you for submitting your manuscript to ICT Express. I have completed the review of your manuscript and a summary is appended below. The reviewers recommend reconsideration of your paper following major revision. I invite you to resubmit your manuscript after addressing all reviewer comments.
When resubmitting your manuscript, please carefully consider all issues mentioned in the reviewers' comments, outline every change made point by point, and provide suitable rebuttals for any comments not addressed.
To submit your revised manuscript:

Log into EVISE at: http://www.evise.com/evise/faces/pages/navigation/NavController.jspx?JRNL_ACR=ICTE
Locate your manuscript under the header 'My Submissions that need Revisions' on your 'My Author Tasks' view
Click on 'Agree to Revise'
Make the required edits
Click on 'Complete Submission' to approve

What happens next?
After you approve your submission preview you will receive a notification that the submission is complete. To track the status of your paper throughout the editorial process, log in to Evise at: http://www.evise.com/evise/faces/pages/navigation/NavController.jspx?JRNL_ACR=ICTE.
Enrich your article to present your research with maximum impact. This journal supports the following Content Innovations:
I look forward to receiving your revised manuscript as soon as possible.
Kind regards,
Professor. Pack
Editor
ICT Express
Comments from the editors and reviewers:
-Reviewer 1

- This paper proposed the offloading framework which determines the task offloading and resource allocation.

I have some questions.
1. Which entity determines the task offloading and resource allocation policy?. I think that since the end device cannot determine the decisions, the controller which determines the decisions is needed.
Please, clarify the system model and the overall procedure. In my opinion, the delay model should be re-defined if the controller is added.

2. According to constraints(14a), alpha_n can be set to one or zero. But according to (14c), alpha_n should be less than F. What is F ? and since if F is set a value less than 1 all alpha_n should be determined as 0, F should be set over 1. Anyway, how (14c) can represent "the edge servers resource constraint"? I cannot understand constraint. Please, clarify these constraints and description.

3. State, action, reward should be formulated to use Q-Learning. What is reward? Please, clarify the Q-learning formulation. In my opinion, it is better to reduce the description of basic Q-learning and add the description of proposed Q-learning based offloading scheme.


In this paper, there are many type errors and inconsistent definition.
For instance,
1. in Related Work Section, rea time data streaming (in the first stance) should be corrected as real time~.
2. in 3.1 System Model, the authors wrote: "The user equipment (UE) set is stand for N = 1, 2, 3,..., N". Is this definition correct?. I think that The UE set will be defined as N \in {1,2,3, ..., n, ..., N_max}.
3. on page 3, alpha_n = 0 means the end device n offload task to edge server, but on page 4, alpha_n = 0 means that n end device executes its task by local computing. Why the definition of alpha is different?
4. "n end devices" was written. Is this correct term? "n end devices" means the number of devices is "n"
Besides these, there are many incorrect definitions. The author should revise the paper by considering these problems.

-Reviewer 2

-
The authors examined the scheme fo a task offloading approach for the edge devices considering IIoT networks with computing resources at the end devices.
The paper is very well written and has provided a list of up-to-date references. The paper is interesting and well organized. However, this paper is needed to revise/update some parts.
Please shows recall traces outperform
Please explain how the use of a backtracking model and recall traces
Please shows proposed model as compared model with Average reward vs. timesteps
Please explain what a good table means in Figure 4
Please explain simulation environment