Lightweight Model for Botnet Attack Detection in Software Defined Network-Orchestrated IoT

Round 1

Reviewer 1 Report

In this paper, the authors proposed a lightweight deep learning model for Botnet attack detection in a Software-defined IoT network. The research is interesting and the paper is well-written. However, to improve the quality my comments are given below.

1) The authors are suggested to modify the paper title. The paper title includes a complete word for SDN. For example; 

Lightweight Model for Botnet Attack Detection in SDN Orchestrated IoT Software-defined networking etc.

2) Please follow the journal template and modify your paper. For example; remove Version March 15, 2023, submitted to Journal Not Specified.

3) The paper organization is not mentioned in the introduction section, Please add; the rest of the paper is organized as follows. In section 1,xxxx

Please cite and follow the below paper(As a model paper) most relevant to your research.

Abbasi R, Mateen A, Ali Abid M, Khan S. A Step toward Next-Generation Advancements in the Internet of Things Technologies. Sensors. 2022; 22(20):8072. https://doi.org/10.3390/s22208072

4) The motivation, benefits, and contribution should be highlighted in bullets please have a look at the above-mentioned paper.

5) Please add the complete details of the datasets and simulation toll that you have used in section 3.1.

6) Improve the quality of figures in the overall manuscript.

7) Future work should be mentioned in the conclusion section.

Author Response

Dear Reviewer, thank you for your feedback and your concerns are addressed.

With best regards,

On behalf of all authors

Author Response File: Author Response.pdf

Reviewer 2 Report

The authors proposed a lightweight deep learning model for an SDN-enabled IoT framework that leverages IoT devices by providing resources to deploy instant protection against the five Botnet malware attacks: DoS, DDoS, Fuzzing, Boofuzz, OS Fingerprinting and Port Scanning. The lightweight model is achieved by carefully designing a deep learning model architecture with four convolutional layers, a few filters, and overall average pooling thereafter. This model proposed by the authors can achieve performance with high accuracy while using less computational resources and solving resource limitation issues. The scientific contribution is well presented. Nethertheless, some remarks should be addressed to improve the quality of the paper:

- Too many typos: for example at line 272, ...

- Words are attached for example in lines 80, 91, 267, 272, 310, ....

- Figure 3 is unclear try to redraw it.

- Figure 4 is not legible. Try to enlarge the figure.

Author Response

Dear Reviewer, thank you for your feedback and your concerns are addressed.

With best regards,

On behalf of all authors

Author Response File: Author Response.pdf

Reviewer 3 Report

This article presents a Lightweight Model for Botnet Attack Detection in SDN Orchestrated IoT and is adequate for a dynamic journal of sustained academic excellence like MDPI Applied Sciences

All the sections of this article are well-written, well-balanced, and well-presented. The authors use descriptive language to present the proposed framework, methodology, analysis, and experimental findings.

In Table 7 the authors present a comparison of the proposed method with previous work on SDN Datasets for 10-IoT devices, please provide us in the text an explanation of how did you perform this comparison? Are you referring to algorithm 3? Table 7 is not referenced in the text.

My minor comments:

Lines187-188: please rephrase the sentence.

All Tables follow mdpi format

Lines 268-281: please check some typos and paragraph formatting (use of the abbreviation eqn etc.)

Refer to all figures in the text with capital F

Improve the quality of Figures 3 and 4, enlarge the figures

Author Response

Dear Reviewer, thank you for your feedback and your concerns are addressed.

With best regards,

On behalf of all authors

Author Response File: Author Response.pdf