Multilevel Task Offloading and Resource Optimization of Edge Computing Networks Considering UAV Relay and Green Energy

Round 1

Reviewer 1 Report

The paper presents an interesting idea using UAV and MEC servers as a platform for offloading computational task using MDP. The results are promising. I have the following comments:

• The authors provided the environment where such architecture can be applied but they did not provide a specific application that needs to be offloaded. This is also applied in the simulation part where more details about the task should be provided. This will strengthen the paper.
• In the text “Given the complexity of the algorithm, decision-making time cannot be ignored. The terminal equipment can use a table to store the best offloading strategy for the tasks computed by the MDP algorithm. This table includes the best decision for each state of the computing task which can be obtained by pre-calculation. Hence, the decision time of this algorithm is short and will not affect the task calculation”. This is should be clarified more. In other words, where we execute the provided algorithm. Is it going to be in the terminal device? Does the terminal run the MDP model. This should be explained more.
• Figures 8 and 9 should be organized in a better way.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Some changes and clarifications need to be performed in the manuscript. These are presented below:

Page 5 – Repetition of a sentence

Page 7 – In equation 7 does the variation with d^-4 is always true no matter what are the relative positions of the terminal and the UAV?

Page 9 – In equation 14, E_uav-c should be the UAV initial energy.

Page 10 – The authors use d_u for a variable that was previously defined as d. The same question posed regarding equation 7 applies to equation 18. Does the exponential decay with d depend on the relative position between the UAV and the G-MEC?

Page 11 – Last line – Replace t_min with t_max.

Page 12 – N_L is wrongly written in equation 36. The authors use K and k indifferently. Please harmonize the use.

Page 20 – Figure 7 b) is missing the variable in the y-axis

Pages 21 and 22 – There is a problem with figures 8, 9, 10 and 11 (formatting and legends)

Page 23 – Different font sizes are used.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

This paper shows the moving MEC server offloading scheme using UAV. In UAV-based moving edge server system, the most important thing is how to appropriately locate the moving edge servers and how to migrate service load between moving edge servers, between moving edge server and G-MEC, etc. However, this paper just shows general cases and performance results, and it lacks any differentiated points when compared to existing related works.

Additional minor points:

• Fig. 8 & 9, 10 & 11 relocation are required
• Overall descriptions are required to be checked again. For example, in page 5, the same sentence is shown in a row.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

This reviewer thinks that this revision reflects well the previous review comments.