kumoh national institute of technology
Networked Systems Lab.

Cosmas, ICTE 2019 366: "Selecting Gaussian Process Regression Kernels for Evaluating MIMO Antenna Configuration"
By : Cosmas
Date : 2020-01-13
Views : 323

Ref: ICTE_2019_366
Title: Selecting Gaussian Process Regression Kernels for Evaluating MIMO Antenna Configuration
Journal: ICT Express

Dear Professor. Kim,

Thank you for submitting your manuscript to ICT Express. I regret to inform you that your paper is not acceptable for publication. We have completed the review of your manuscript and a summary is appended below. The reviewers have advised against publication of your manuscript and I must therefore reject it at this time. For your information and guidance, any specific comments explaining why I have reached this decision and those received from reviewers, if available, are listed at the end of this letter.

You have the option of resubmitting a substantially revised version of your paper, which would be considered as a new submission. If you decide to do this, you should refer to the reference number of the current paper and include a cover letter which explains in detail how the paper has been changed or not, in reply to the Editor and Reviewer comments.

Thank you for giving us the opportunity to consider your work.

Kind regards,

Professor. Ju
Editor
ICT Express

Comments from the editors and reviewers:
-Reviewer 1

- This paper investigated the exponential Gaussian process regression (EGPR) for training and predicting antenna configuration in MIMO system of 5G for improved capacity utilization. The interesting point was to use Machine Learning tool of Matlab to predict performances, but the comparison results with other schemes were not clear and the most of results are straightforward, not interested. In my opinion, this paper needs to show more clear and detail descriptions, particularly in comparison parts. Overall, the contribution of the paper is some limited in general. I hope that the editor can make a fair decision with other reviewers comments.


-Reviewer 2

-
In this paper, the authors have proposed to adopt the exponential Gaussian process regression (EGPR) as the kernel for the performance evaluation of MIMO antenna configuration. However, the following comments may give some help.



- At the 5th line of the second paragraph of Section 1: was -> were



- At the 23th line of the second paragraph of Section 1: has -> have



- At the 22th line of the third paragraph of Section 1: Fig.[1] -> Fig. 1.



- At the 1st line of the first paragraph of Section 3: Gaussian process regression(GPR) -> GPR



- Please clarify the sentence at the 3rd line of the second paragraph of Section 3: computing the ergodic capacity (C) or spectral efficiency will by equation (6)



- Please clarify the sentence at the 6th line of the second paragraph of Section 3: s is the scaling factor which is a function of the ratio of Tx and Rx and signal-to-noise ratio(SNR).



- Please specify what are Ai and Pi at (7).



- As shown in Fig. 2 and Fig. 3, there are very high error margins. For example, when the number of transmit antenna is 5, there is 90% error at Fig. 2. Therefore, it seems that the GPR prediction is not suitable for performance evaluation of MIMO antenna configuration. The errors in Fig. 2 and Fig. 3 are tolerable to evaluate the MIMO antenna configuration?