Next Article in Journal
Quad-Beam 4 × 2 Array Antenna for Millimeter-Wave 5G Applications
Next Article in Special Issue
Distributed AI-Driven Simulation Framework for Performance Evaluation of Hybrid Satellite–Terrestrial Network Access
Previous Article in Journal
Generative AI in Education: Perspectives Through an Academic Lens
Previous Article in Special Issue
On Adaptation of Resources in New Radio Vehicle-to-Everything Mode 1 Dynamic Resource Allocation
 
 
Article
Peer-Review Record

Hybrid Long-Range–5G Multi-Sensor Platform for Predictive Maintenance for Ventilation Systems

Electronics 2025, 14(5), 1055; https://doi.org/10.3390/electronics14051055
by Praveen Mohanram 1,* and Robert H. Schmitt 2
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Electronics 2025, 14(5), 1055; https://doi.org/10.3390/electronics14051055
Submission received: 26 January 2025 / Revised: 28 February 2025 / Accepted: 3 March 2025 / Published: 6 March 2025
(This article belongs to the Special Issue 5G Mobile Telecommunication Systems and Recent Advances)

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The presented paper is devoted to the issues related to the presentation of the IIoT technology for use in controlling the ventilation system. The system is described in quite detail from the construction side. All of its features are also presented in an illustrative way, as well as issues related to security and information exchange techniques. The only research element concerns simple measurements and calculations regarding energy consumption using different operating modes of the proposed hybrid system with LoRa/3GPP 5G wireless communication. The article was written in an accessible way, using a fairly simple specialist language, which will certainly increase the number of readers. The obtained results of the analysis and calculations are very simple and constitute a very small part of the paper. In this form, the paper is rather a technical description of the equipment. In order to improve the quality of the paper content, as a scientific one, it is recommended to introduce several corrections and significant extensions:

  1. The last sentence of the Abstract: The hardware architecture has been described very extensively. Data analysis has been practically not presented, only general encyclopedic descriptions we can find in the paper. The software for ventilation control with prediction has also not been presented, so this part should be expanded.
  2. The performance of the presented hardware platform should be compared with others available on the market, not only through references to literature, but through a physical or mathematical comparison of energy and transmission capabilities. In this respect, one cannot rely only on recently published articles by the authors.
  3. How does the power consumption depend on the length of the propagation path? In this respect, no specific link parameters (MCS diagram) have been mentioned in either the LoRa interface or 3GPP 5G.
  4. A diagram or algorithm for exchanging data/information via the LoRa and 5G interfaces (especially for integrated IoT LWM2M and OTA techniques) should be drawn so that the time occupancy used to calculate the power consumption can be easily traced.
  5. Other minor detailed comments:
    1. Line 9: duplicate word
    2. Line 36: it should be „Long Range (LoRa)
    3. Figure 1: it should be “Feature Extraction”
    4. Line 493: reference number is missed
    5. Lines 693-701: too big font

Author Response

Thank you for the detailed reviewed

Comment 1 : The last sentence of the Abstract: The hardware architecture has been described very extensively. Data analysis has been practically not presented, only general encyclopedic descriptions we can find in the paper. The software for ventilation control with prediction has also not been presented, so this part should be expanded.

Response : Yes the software for ventilation control and  data analysis is not presented as its not the focus of the paper.

Comment 2 : The performance of the presented hardware platform should be compared with others available on the market, not only through references to literature, but through a physical or mathematical comparison of energy and transmission capabilities. In this respect, one cannot rely only on recently published articles by the authors.

Response : A comparison of the currently available hardware is compared on based on mathematical analysis and included. 

comment 3 . How does the power consumption depend on the length of the propagation path? In this respect, no specific link parameters (MCS diagram) have been mentioned in either the LoRa interface or 3GPP 5G.

Response : I agree on the comment. While the focus on the hardware design, the detailed discussion for MCS and power consumption is now added in a small section to provide some consideration for power consumption. 

 Comment 4 :A diagram or algorithm for exchanging data/information via the LoRa and 5G interfaces (especially for integrated IoT LWM2M and OTA techniques) should be drawn so that the time occupancy used to calculate the power consumption can be easily traced.

response : The software functionality is described and a software architecture is preseneted, a new diagraom for data/information flow is added to make it easier for analysis.

Comment 5 : Other minor detailed comments:

    1. Line 9: duplicate word
    2. Line 36: it should be „Long Range (LoRa)
    3. Figure 1: it should be “Feature Extraction”
    4. Line 493: reference number is missed
    5. Lines 693-701: too big font

Response : all are fixed.

Reviewer 2 Report

Comments and Suggestions for Authors

Summary

This paper presents a hybrid LoRa-5G multi-sensor platform for predictive maintenance in ventilation systems. By integrating LoRa’s energy efficiency with 5G’s high-speed connectivity, the system aims to achieve real-time monitoring while optimizing power consumption. The ventilation system is monitored through a variety of different sensors, including vibration, temperature, humidity, power measurement, and air quality. The sensor nodes then run pre-trained TinyML models to infer the status of the machinery.

The key idea is to use a hybrid communication strategy where LoRa transmits routine low-power data/inference results, while 5G handles periodic high-bandwidth updates.

The system performance is evaluated by comparing LoRa-only, 5G-only, and hybrid setups, showing that the hybrid approach reduces power consumption while maintaining real-time data transmission.

Strengths

  • The authors introduce an innovative hybrid communication model that successfully combines LoRa (low-power) and 5G (high-speed) to balance energy efficiency and real-time monitoring.
  • The paper demonstrates a well-structured validation methodology by comparing hybrid, LoRa-only, and 5G-only configurations demonstrating relevant improvements in power consumption and data accuracy.
  • The paper provides a detailed hardware and software design, outlining sensor integration, power management strategies, and system architecture.
  • The authors address security and scalability by discussing data encryption and over-the-air (OTA) updates.

Weaknesses

  • The evaluation is limited, as validation is limited to lab experiments without field tests in actual industrial environments.
  • The paper omits a cost analysis, making it unclear whether the hybrid system is economically viable for large-scale industrial adoption.
  • The authors lack a detailed discussion on AI model accuracy and on the dataset used for training or analyze the impact of environmental variations on model performance.
  • The paper lacks a dedicated section comparing this study to existing research. The authors should cite relevant studies to better position their contributions within the existing literature. Among some relevant research work:
    • Chen et al. (2024). "LoRa mesh-5G integrated network for trackside smart weather monitoring." IEEE Transactions on Industrial Informatics.
    • Busacca et al. (2024). "FedLoRa: IoT Spectrum Sensing Through Fast and Energy-Efficient Federated Learning in LoRa Networks." IEEE 21st International Conference on Mobile Ad-Hoc and Smart Systems.

 

Author Response

  • The evaluation is limited, as validation is limited to lab experiments without field tests in actual industrial environments

Response : The hardware was designed for the industrial environement, and the analysis for power measurment would not vary . The data analysis needed is not the focus

  • The paper omits a cost analysis, making it unclear whether the hybrid system is economically viable for large-scale industrial adoption.

Response : cost analysis is now included

  • The authors lack a detailed discussion on AI model accuracy and on the dataset used for training or analyze the impact of environmental variations on model performance.

response : This is work in progress and the data acquisition is ongoing, we expect the AI model accuracy will  be evaluated in another paper. But the use of AI model such as LSTM, requies dataset which is currently not present.

  • The paper lacks a dedicated section comparing this study to existing research. The authors should cite relevant studies to better position their contributions within the existing literature. Among some relevant research work:
    • Chen et al. (2024). "LoRa mesh-5G integrated network for trackside smart weather monitoring." IEEE Transactions on Industrial Informatics.
    • Busacca et al. (2024). "FedLoRa: IoT Spectrum Sensing Through Fast and Energy-Efficient Federated Learning in LoRa Networks." IEEE 21st International Conference on Mobile Ad-Hoc and Smart Systems

Response : there is a new section where there is comparison of study.

Reviewer 3 Report

Comments and Suggestions for Authors

This paper describes the hardware architecture, data analytics, and software design used to perform predictive maintenance for a ventilation system. The paper includes interesting content, but the manuscript is somewhat in the form of a report, so it needs to be revised for publication. 

(1) It is desirable to explain in more detail the research trends and industrialization trends of the multi-sensor platform, LoRa IIOT system, and 5G-LoRa platform in the introduction. 

(2) It is necessary to explain in detail the contribution of the paper in the introduction. 

(3) It is desirable to describe the organization of the paper in the introduction.

(4) In the introduction, it is necessary to explain the motivation for using LoRa together with 5G-mMTC, or the pros and cons of using them together. 

(5) Section 2 is about methodology, so it is necessary to explain the environment under consideration and the problem to be solved before that. 

(6) Abbreviations such as NB-IoT, BMS, IC, USB, PCB, etc. should be defined first and then used. 

(7) In the abstract, LoRa, etc. should be defined first and then used. 

(8) Lora Range is an incorrect expression. 

(9) It is not desirable to define the same abbreviations such as LoRa, OTA, etc. multiple times. 

(10) It would be good to mention the cost aspect.

(11) The table should have a table number.

(12) It is also necessary to explain whether there are any disadvantages to the hybrid approach.

(13) What is (ref) in "For the ventilation unit monitoring application, the 295
Quectel 5G M.2 RM520N module is used (ref)"?

 

Author Response

(1) It is desirable to explain in more detail the research trends and industrialization trends of the multi-sensor platform, LoRa IIOT system, and 5G-LoRa platform in the introduction. 

Response : The introduction is now modified to include the information

(2) It is necessary to explain in detail the contribution of the paper in the introduction. 

Response : The introduction is now modified to include the information

(3) It is desirable to describe the organization of the paper in the introduction.

Response : The introduction is now modified to include the information

(4) In the introduction, it is necessary to explain the motivation for using LoRa together with 5G-mMTC, or the pros and cons of using them together. 

Response : The introduction is now modified to include the information

(5) Section 2 is about methodology, so it is necessary to explain the environment under consideration and the problem to be solved before that. 

Response : The Methodology is now modified to include the information.

(6) Abbreviations such as NB-IoT, BMS, IC, USB, PCB, etc. should be defined first and then used. 

Response : Abbreviations are defined first

(7) In the abstract, LoRa, etc. should be defined first and then used. 

Response : Its defined in abstract now.

(8) Lora Range is an incorrect expression. 

Response : modified.

(9) It is not desirable to define the same abbreviations such as LoRa, OTA, etc. multiple times. 

response : modified and removed the redundant abbreviations.

(10) It would be good to mention the cost aspect.

Response : The cost aspect is included

(11) The table should have a table number.

Response : its modified.

(12) It is also necessary to explain whether there are any disadvantages to the hybrid approach.

response :  a small information is added for disadvantage.

(13) What is (ref) in "For the ventilation unit monitoring application, the 295
Quectel 5G M.2 RM520N module is used (ref)"?

response : reference addded

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Almost all the reviewer comments included in the previous version of the peer-review were taken into account. Thank you for introducing corrections and explanations to the paper in accordance with the reviewer's recommendations. I have no further comments.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have addressed all the Reviewer's comments. The paper can now be accepted for publication.

Back to TopTop