Radio Coverage and Device Capacity Dimensioning Methodologies for IoT LoRaWAN and NB-IoT Deployments in Urban Environments
Round 1
Reviewer 1 Report
The authors present coverage and capacity capabilities of LoRaWAN and NB-IoT. The paper includes experiments, however, the experimental scenarios are not described at all. A major revision is required.
C1: Introduction. Features are presented of characteristics of Wireless technologies in a form of mixed metrics with characteristics. Please identify which is which. Is throughput a metric or a characteristic? Are there any corresponding metrics for Energy Efficiency? Sensing capabilities? Mobility capabilities? how these attributes affect wireless communications of LoRaWAN and NB-IoT? Is BW a metric, is capacity a characteristic? What about ToA, receiver sensitivity and duty cycle? Please provide an additional Table that connects attributes-characteristics with metrics. What about security?
C2: section 2 LoRaWAN vs NB-IoT. Several technologies are presented. Please revise title to existing wireless technologies to fit with the available content. Why LTE-M and not IEEE802.15.4? Please provide a cross-comparison table of all available IoT technologies based on attributes( NB-IoT, LTE-M, LoRAWAN, ZigBee, BLE)
C3: Section 3- What is actually presented here? How does Figure 1 fits to section 3? Is it a generic framework? Is it actually the experimental methodology? Please use LoRaWAN instead of LoRa.
C4: Please provide a section that describes your experimental methodologies and performed scenarios
C5: LoRaWAN classes A,B,C, please mention the different use of the classes
C6: Table 1 and Table3 LoRaWAN/NB-IoT propagation model OUTPUT. I think it is the distance from the gateway or cell(concentrator) and not maximum range
C7: The LoRaWAN and NB-IoT models shall precede the Experimentation methodology and scenarios, as a theoretical approach
C8:Figure 2 : Please mention of Limitations of LoRaWAN regarding ToA. Please mention available frequencies and channels of NB-IoT
C9: Figures 4/8. Regarding capacity. How this experimentation has been performed? Is it something taken out from bibliographic research?
C10:Figures 2/5. Please check if ToA results of LoRaWAN and and NB-IoT are correct.
C11:Please provide a cross-comparison table between LoRaWAN and NB-IoT with respect to ToA, Throughput, capacity over distance (low, medium, high) and a short discussion subsection of the results.
C12: Please also check:
https://ieeexplore.ieee.org/abstract/document/8088231
https://www.mdpi.com/1999-5903/9/4/78
https://www.mdpi.com/2411-5134/4/3/52
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The authors propose a review of the IoT network deployments in licensed and unlicensed bands.
It is not clear the contribution of this paper in front of state-of-the-art. It is necessary to create a section for related work.
I suggest the authors reconsider reducing the title's size.
All equations need to be rewritten, not paste an image.
In the present form, the paper appears to be a unification of third-party models. It is necessary to improve the section of the model proposed to justify the model's confidence.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
**Reviewers Recommendation**
Major Revision
**Summary**
The authors present the problem for accurate planning and modeling tools in the scope of massive network deployments based on IoT devices where the wireless medium presents a key resource.
Next, the authors present the background related to the heterogeneity of IoT scenarios, regarding several characteristics for the network bandwidth usage and power efficiency related to link budget and reliability.
Then, two popular LPWAN technologies, NB-IoT and LoRaWAN, have been introduced and analyzed.
Finally, the authors provide a planning model for achieving networking scalability and capacity in massive IoT scenarios, based on LoRaWAN and NB-IoT.
**General Comments**
The manuscript is missing a proper related-works/state-of-the-art section. Only the last paragraph of section 2 briefly covers other works related to modeling scalable networks. And even in that case, the authors heavily leverage on those works to develop their own model.
The cited works seem to be old. Averaging years 2017-2018. This is odd as it seems that IoT and LPWAN networks are a strong topic in both academia and or industry.
There distinction between LoRa PHY and LoRaWAN is not clear. It seems in the work that those two things are the same when in fact they're are different network stack layers.
For instance, there are some key documents regarding LoRa/LoRaWAN not cited in the present manuscript. These works include:
- SX1272/3/6/7/8 LoRa Modem Design Guide, AN1200.13
- LoRa Modulation Basics AN1200.22
- LoRaWAN ™ Specification v1.0.2
- LoRaWAN™ 1.0.2 Regional Parameters
The contents in Section 3.A LoRaWAN Model seems to be taken almost verbatim from the most important references regarding the LoRa PHY technology, namely
- SX1272/3/6/7/8 LoRa Modem Design Guide, AN1200.13
- LoRa Modulation Basics AN1200.22
These documents have been already studied and discussed in previous LoRa/LoRaWAN scientific works ad-nauseam, therefore, duplicating this information in the manuscript is not justified.
The LPWAN technologies discussed behave differently regarding the Regional Area being used. This affects Duty Cycle, and other PHY parameters related. There seems to be no differentiation regarding the regional area discussed in the work and this is an important matter.
The contents of several tables and figures in Section 3 and 4 are somewhat repeated and are hard to justify since it seems like they give away very similar information. For instance Tables 1 and Table 6 could be merged into one.
In Table 1, the maximum coverage ranges seem outdated, since there are lots of works in the literature that have proven much larger distances in coverage results.
The chosen scenario where one byte is transmitted is not typical, specially when LoRaWAN and NB-IoT have severe network overhead due to mandatory network stack headers, where there is a high overhead due to the mandatory 13 byte header in LoRaWAN as well as a minimum 5 bytes in NB-IoT. This scenario decision has not been properly studied and or justified.
**Suggested Improvements**
1)
Equations and figures are hard to read, even zooming in. Please use more picture quality.
2)
Regarding the scenario of transmitting one single byte, please consider that LoRaWAN actually introduces a mandatory 13-byte header in each uplink communication, as well as NB-IoT introduces a mininum of 5 bytes in each transmission. Please redo your calculations and update your model results in order to reflect on these facts. Additionally, further clarify or justify why sending one single byte of user payload is a viable strategy.
3)
Line 241, [12] is referenced, when in fact it is a NB-IoT related citation. Please check.
4)
Please further clarify the distinction among LoRaWAN and LoRa PHY layers. When you specify LoRaWAN, you should use the correct terminology presented in the LoRaWAN ™ Specification v1.0.2 document, and or LoRaWAN™ 1.0.2 Regional Parameters. Likewise, when talking about LoRa, a different set of terms could be employed, such as those present in LoRa Modem Design Guide, AN1200.13.
5)
In Table 1, please consider reviewing other works in the literature to update the figures of maximum coverage area since it seems like the presented distances are way too small. Alternatively, in the Table caption, please indicate the source work where this information has been obtained.
6)
Please consider merging the contents of Table 1 and 6 since it seems that another additional tables could be avoided. Please review the rest of figures and or tables and consider merging when possible, since it seems like they share a lot of common parameters. Alternatively, please further clarify why the contents of these tables, which share columns or variables, could not be merged.
7)
The contents in Section 3.A LoRaWAN Model seems to be taken almost verbatim from the most important references regarding the LoRa PHY technology, namely
- SX1272/3/6/7/8 LoRa Modem Design Guide, AN1200.13
- LoRa Modulation Basics AN1200.22
Please, try to be more concise about the information presented in those technical documents and, since the information is well known by LPWAN experts and does not contribute to the authors results. Similarly, apply the same procedure to NB-IoT model section, since it has also been presented ad-nauseam in the literature.
8)
Please, perform a more in-depth literature review of the state-of-the-art research proposal regarding the planning and modeling of massive IoT networks. Also further clarify or specify how the contributions of this paper contrasts next to other perform reviews. For instance, maybe indicate with a YES/NO Table what are the contribution features of other state-of-the-art research proposals or solutions regarding IoT network planning and modeling tools VS the one presented in this work.
9)
Please review and include citations to the following items where appropiate:
- SX1272/3/6/7/8 LoRa Modem Design Guide, AN1200.13
- LoRa Modulation Basics AN1200.22
- LoRaWAN ™ Specification v1.0.2
- LoRaWAN™ 1.0.2 Regional Parameters
10)
The cited works seem to be old. Please give a brief description of what paper databases were employed (e.g. web of science), and what keywords where employed during the literature review in order to justify the cited works' years of publication.
11)
Please consider including real-life deployment data in order to validate your proposed model results in section 4.
12)
Please further clarify or explain what Regional Area are you considering (e.g. EU868, USA, ASIA) for the methodology and results regarding LoRaWAN and NB-IoT, since they behave differently depending on the global region.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
None
Author Response
Please refer to the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
The authors improved the paper since the last version and can be accepted in the present form.
Author Response
Please refer to the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
First, I would like to congratulate the authors for the work put in this new revision of the manuscript.
However, it is in this reviewer's opinion that Comment 8) was not successfully addressed by the authors.
In the new manuscript, lines 111--203 could be considered part of the background required to understand the work contributions, while the rest of section 2 is the real related works section. Please consider moving this information to a different section of the work. E.g. Section 1, 3, or a whole new section.
Still, I feel like the proposed contribution is not clearly contrasted against the cited related works, as stated in the original Comment 8).
Please include a full paragraph at line 236 stating what other contributions cited (12--20...) lack or are missing, and that is has been successfully achieved by the presented work. For instance, consider talking about the weaknesses or missing gaps found in the cited works in contrast with the strengths or contributions of the presented paper in a summary.
Author Response
Please refer to the attachment.
Author Response File: Author Response.pdf
