Spectrum Demand Forecasting for IoT Services

Round 1

Reviewer 1 Report

-The paper should be interesting ;;;
-block diagram of the research should be added step by step;;;;
-Figures should have high quality, TIFF format;;;
-what will society have from the paper?;;;
-what is the result of the paper?;;;
-please add photos of the application of the proposed research, 1-2 photos ;;;
-formulas and fonts should be formatted;;;;
-please add labels + SI units if any etc. 
-references should be new (2019-2021 web of science), 50% of references

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The authors present a methodology for spectrum forecasting consisting of two phases: 2 a market study and a spectrum prediction model. The work is interesting and the proposed methodology is adequately explained. 

However, the authors should enhance a bit the motivation behind this work in the abstract and the introductory section.

Moreover, the contributions of this paper compared to other relevant works are not clear. To this end, the authors should include a new section called "Related Work" where similar works can be described and the differences can be highlighted. Can your approach overcome other machine learning solutions?

The market study is described adequately. However, a figure can be included, providing its overview. 

The experimental results demonstrate the efficiency of the work. It would be interesting if the author can include optionally a comparative analysis with respect to their forecasting model.

Finally, the paper should be re-checked entirely for possible typos and writing errors.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The authors have proposed an interesting methodology to forecast the spectrum demand and usage for IoT and 5G communication scenario. It is a very interesting  topic of research, especially when the IoT devices are increasing exponentially. The proposed scheme is evaluated for Colombian region, which gives this manuscript a national level importance and real-time statistics. I recommend to revise this paper and address the following concerns. 

1. The authors haven't compared or discussed the AI based spectrum forecasting methods. It is interesting to see the a comparison of the complexity of the algorithm and the achieved results.
2. What change authors will recommend in the proposed algorithm, so that it is applicable globally rather than limited to a national level implementation. 
3. Most importantly, the authors have forecasted the spectrum allocation and usage till 2030. But they haven't discussed the 3GPP evolution for 5G, IoT or M2M communication. The 3GPP v17 is already in place where as 3GPP v18 has a release date of 2022. The new version may change the whole dynamics of the spectrum allocation strategies. It is nice to address it in the manuscript.   

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

The authors addressed most of the comments sufficiently. Therefore, the paper can be accepted for publication. It would be beneficial whether the authors can improve the evaluation analysis with a more detailed comparative study.

Reviewer 3 Report

The authors have revised the manuscript and addressed my concerns. I recommend to accept the article.