Grapevine Downy Mildew Warning System Based on NB-IoT and Energy Harvesting Technology
Round 1
Reviewer 1 Report
This is a research article in which the authors propose a grapevine downy mildew warning system based on NB-IoT and energy harvesting technology. The authors claim their model utilizes more parameters, which allow their model to be more accurate than the traditional ones. They performed a comparison with a commercial iMetos warning system for three selected vineyard locations demonstrating a correlation with the commercial system.
Comments:
Generally, please try to reference Figures in text before the actual figure position (if possible).
In the abstract, please be more specific on the additional parameters used in the proposed model.
In Introduction, the readers would benefit from a better and more detailed explanation of Figure 1.
In Section 3, please be more specific on the component selection and their characteristics, especially for the sensors used.
Also, Figure 2 is not very clear. What are the five icons below? Please consider using appropriate UML diagrams (i.e. UML Deployment and/or Component diagrams).
Figures 3 and 4, it would be beneficial to add more description for both algorithms, and also to emphasize the novelty of these algorithms as opposed to the ones described in the literature review.
Figure 5, please consider adding a block diagram to the right of the photo.
Figure 8, font is too small, illegible.
Discussion related to Figures 13-18, could be improved
In addition, it would be good to have a table with a list of functional and non-functional requirements.
Table 3, It would be beneficial to have a legend here next to the table. Also, it would be good to quantify the alignment between the proposed system and iMetos in %% or with some other metrics.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
It is an interesting work to develop a early alert system for vineyard. Authors have spent most paragraphs to describe the rules for the detection and prediction, but no their own contribution on the algorithm for data analysis. As authors mentioned in the paper, no operation record available to valid the performance of proposed algorithm. Regarding the system integration, it more like a piece of work in technology development. It may be more appropriate to appear as a letter or communicate for an advertisement for a new product.
As a research paper, authors shall make more revision as following:
- Real infection cases should be used to validate the proposed method, and comparison with commercial software warning results alone cannot validate the effectiveness of the proposed method.
- A comparison with the methods proposed in recent papers is required. For example, multivariate based methods, image recognition based methods…
- The block diagram of the algorithm is not clearly expressed and the authors need to reorganize the algorithm logic. For example, authors do not explain the meaning of "hour" in Figure 3 and Figure 4, and the meaning of "Reset 3-10" in Figure 3.
- Figure 11 is not mentioned in manuscript.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
This study proposes a new model and early warning system for grape downy mildew based on NB-IoT and energy harvesting. This model was developed using additional parameters, which are more accurate than traditional models. And compared with the actual situation, the reliability of the work is verified. The specific issues are as follows: (1) In the abstract of the study, the background part is too cumbersome and wordy. It is recommended to simplify the background and introduce the work of this sutdy as soon as possible; (2) The results of the study are compared with commercial iMetos in three specific vineyard locations. The warning system was compared, showing a significant correlation between the alarms. However, the results obtained by such a comparative design cannot prove that the application effect of this study is better. (3) The Literature Review of the second section should be merged with the introduction of the first section, and the research background should be simplified. A lot of information irrelevant to this study should be deleted; (4) There are so many grammar and expression errors, for example, line 11, "which we show to be more accurate" , This is a wrong statement; (5) The results and discussion section in the study does not actually discuss of the content. It is recommended to add a section of real discussion, from the design of the article, the results obtained and conclusions, etc., to compare the work of the study with the previous work to demonstrate the advancement and limitations of the work.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
This is a revised version of the research article in which the authors propose a grapevine downy mildew warning system based on NB-IoT and energy harvesting technology. The authors claim their model utilizes more parameters, which allow their model to be more accurate than the traditional ones. They performed a comparison with a commercial iMetos warning system for three selected vineyard locations demonstrating a correlation with the commercial system. There is a substantial improvement and the authors responded to all of my questions and concerns.
I recommend the acceptance of the manuscript, with suggestions to implement the minor improvements:
- Please proofread the manuscript for English style and grammar (you may want to reduce the use of “we did this...”, “their model, our model...” and use the passive voice where possible)
- Figure 2, please consider adding text under the icons, in the same fashion as in Figure 1
Reviewer 2 Report
Authors have made changes according to reviewer comments, except the validation by actual data. The paper can be a fair publication for reader.
Reviewer 3 Report
Accept
