Supervised Machine Learning Tools and PUF Based Internet of Vehicles Authentication Framework

Round 1

Reviewer 1 Report

 

The manuscript has many concerns that should be carefully addressed as follows:

The paper is well written. However, it is not well organized. Many subsubsection are not needed with too long descriptions as in Section 4. The discussion of the different types of attacks should be mentioned earlier.

The section numbering mentioned in a descriptive paragraph at the end of the introduction is in Latin. However, afterward, the section numbering is in Arabic.

It is recommended to add a table of abbreviations.

Many notations are not defined. They should be defined directly after their equations.

Fig. 6, 7, and 8 should be clearly discussed.

What is the node type?

What is/are the environment type(s) that the model is tested through?

What is the topology adopted in the proposed model?

The authors should test their model with different data packet sizes.

It is recommended to clearly describe the data processing for the training step. 

More visual results are strictly desired. The authors should present several figures to show the effectiveness and robustness of the proposed model such as accuracy and loss curves comparison, F1-score, IOU, ROC, complexity, precision-recall, and definitely confusion matrix.

The authors are recommended to discuss how to handle the non-linearity problem.

The authors are recommended to discuss how to optimize the hyper-parameters.

The authors should do more surveys on the related works that handled the crucial raised comments. The following papers are recommended to discuss that can assist the authors in this regard:https://doi.org/10.1109/JIOT.2021.3114420; https://doi.org/10.1109/ACCESS.2021.3076119; https://doi.org/10.1109/JIOT.2020.2996671;

It is recommended to add a table of notations.

x-labels of Fig. 11 and 12 are not well located.

It is recommended the authors to discuss each obtained result by a separate paragraph.

Author Response

Thank you for your valuable feedback. We have incorporated it as per our best understanding. Please have the attached file.

Author Response File: Author Response.pdf

Reviewer 2 Report

-Introduction should be revised: (1) Introduce the problem (2)discuss about some of the existing solutions (3)identify the gap or scope of improvement (4) discuss in order to address the identified gaps what is the methodology used (5) list out the contributions (6) Organization of paper.

-Related works section can be enriched with recent works. Authors can refer to the following works:

CANintelliIDS: Detecting In-Vehicle Intrusion Attacks on a Controller Area Network using CNN and Attention-based GRU

An enhanced approach for three factor remote user authentication in multi - server environment

Anomaly Detection in Automated Vehicles Using Multistage Attention-Based Convolutional Neural Network

Robust Authentication System with Privacy preservation of Biometrics

-Figure 6. Schematic view of the proposed authentication framework   should be re-drawn to provide better understanding

-Figure 9. Experimental Setup of the proposed protocol  should be recaptured to make it more understandable

-All equations should be numbered and explained in detail.

-Results should be analyzed in detail and should be compared with an existing approach.

-Conclusions should be revised to include major findings.

Author Response

Thank you for your valuable feedback. We have incorporated it as per our best understanding. Please have the attached file.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have improved the manuscript. I would recommend acceptance of the manuscript in its present form.

Reviewer 2 Report

Authors have addressed all the comments.