SDN-Enabled FiWi-IoT Smart Environment Network Traffic Classification Using Supervised ML Models

Round 1

Reviewer 1 Report

This article presents a Machine Learning supervised network Traffic Classification Scheduling Model in SDN Enhanced-FiWi-IoT. I can see that the authors have done plenty of evaluation work. However, it still has some drawbacks.

• The proposed QoS mapping has not been explained in detail.
• The quality of the figures is very poor.
• The  novelty of the paper is not enough. The techniques used in the paper are actually not very novel.
• The assumptions in the paper are not very clear.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report

The authors suggest a supervised machine learning model for network traffic classification scheduling, in SDN enhanced-FiWi-IoT, that can learn and guarantee the traffics based on its QoS requirements (QoS-Mapping).  Α robust IoT device classification framework is developed, using network-level attributes, e.g., port source and destination, IP address to classify IoT and non-IoT devices.
The manuscript has been written in an elaborated manner. It is well organized and, its structure is appropriate for a research article. The English language needs corrections. There are several typos and grammatical errors that require review.
The Introduction section gives enough information about the topic. Also, the aim and the major contributions of the article are stated clearly.
It is worth noting that the article provides the readers with the necessary background about the Proposed Software-Defined-FiWi-IoT System Architecture and Operation Network Traffic Classification Techniques. It is complete and self-contained.
The review of the state-of-the-art works is supported by some (not enough) recent references.
The research is well designed and, a clear objective has been set. The theoretical model is captured in an elaborated and focused manner and supported in the literature.

The authors give enough details regarding their approach step-by-step. The proposed Machine Learning methodology and the corresponding architecture for the traffic classification process are depicted in Figure 4, which contains the following functional blocks: packet capture and collected, pre-processing and transformed data, ML-training, ML-testing, ML-classification model, and classification results.
The authors have included necessary information about the performance metrics and the environment of the experiments.
The experiments demonstrate the effectiveness of the suggested method. The results are clearly illustrated. 
Although the authors have made a solid work, some points need improvement:
1) The paper title has not the appropriate length; it exceeds ten words. 
2)The abstract should mention the ML model that has the best classification performance and the corresponding numerical results of the performance metrics.                                                                                                              3)In Performance evaluation, the authors should refer to the dataset size and if any preprocessing has been made on it. Is it balanced?
4) The authors should give technical details about the ML models and their configuration under the investigating topic.
5) A discussion section (before the conclusion) is missing. The discussion part needs to provide a comparative analysis with previous studies and be technical. Also, in the same section, the authors should emphasize the limitations and the potential issues of this research.
6)The conclusion section could be improved to highlight the simulation results.                                                                                                                7) The literature should be enriched. Also, reference [2] is not appropriate for a research article; please replace it.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors have addressed my concerns.

Reviewer 2 Report

I have no additional comments on the revised version. The authors addressed all of my concerns.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.