Fine-Grained Identification for Large-Scale IoT Devices: A Smart Probe-Scheduling Approach Based on Information Feedback
Round 1
Reviewer 1 Report
This paper mentions that they have reviewed 58000 IoT devices but not enough evidence or the methodology proposed is not clear.
We recommend all data-set to be opensource in order to be able to validate their methodology.
The conclusion section need a large enhancement. The Literature needs an extension to mention other studies in the area.
Author Response
Please see the attachment
Author Response File: 
Author Response.docx
Reviewer 2 Report
This paper proposes a fine-grained probe-scheduling approach based on information feedback for large-scale IoT devices.
They implemented a prototype system and conducted resl world experiments to validate identification capability. This was evaluated with results to evaluate the performance of their system and also provided a comparison with previous work.
There are a few grammatical and spelling errors across the paper. I would ask the authors to carefully proofread a resubmission.
1. What is the main question addressed by the research? The main question addressed in this paper is the identification granularity of Large-Scale IoT Devices. This paper proposes a fine-grained identification using a smart probe-scheduling approach based on information feedback. 2. Do you consider the topic original or relevant in the field? Does it
address a specific gap in the field?
The topic is original and addresses a specific gap in the identification of large-scale IoT devices.
3. What does it add to the subject area compared with other published
material? Previous work have considered have either used operating system identification or traffic data analysis or device attributes, or 2 of the afore mentioned. This paper proposes the use of three layers: Port, Protocol-response, and Web-Feature layer. The paper further proposes to exploit the relationship between these layers in a framework.
4. What specific improvements should the authors consider regarding the
methodology? What further controls should be considered? The authors should move all conclusion statements into the conclusion sections. This makes it easier for the readers of the work. Is table 2 a list of total devices and then the number is scaled up to a ten fold or is it just a sample list from the 53000 devices probed? This needs to be made clearer.
What about future work and any other open areas?
There are a couple of grammatical errors across the paper that needs to be addressed.
5. Are the conclusions consistent with the evidence and arguments
presented and do they address the main question posed? The conclusions are consistent with the evidence and arguments presented and the main question is addressed. However, I noticed another conclusion section in 3.4. I feel there should be only one conclusion section. 6. Are the references appropriate? References seem appropriate 7. Please include any additional comments on the tables and figures. Tables and figures are legible.
Author Response
Please see the attachment
Author Response File: Author Response.docx
Reviewer 3 Report
This paper seems to give an incremental improvement in the discovery problem for a heterogenous IoT environment. The implementation seems to suggest improved speeds over existing methods.
Author Response
Thanks for your suggections. I have revised my paper based on the suggestions. Please see the attachment
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The response provided by the authors are not sufficient. I can mention that I raised one million devices but if we don't have clear proofs of this it is becoming hard to approve.
Additionally the authors are saying that a firewall is an IoT device? What is the details of the network that has 53000 Iot Devices. Who provided access ? What was the state of the IoT devices when he was running the experiments.
There are a lot of questions that need full details in order the statement 53000 IoT devices to be proved.
