Gait Recognition Algorithm of Coal Mine Personnel Based on LoRa

Round 1

Reviewer 1 Report

The work important in terms of the problem of identification and localization of people in difficult environmental conditions. It contains a large contribution of novelty. It logically presents the methodology for solving the problem of identifying moving objects (in this case people) and the results obtained. However, after analysing the content of the work, a slight modification of the title is proposed to:

Presence (or Localization) Recognition Algorithm of Coal Mine Personnel Based on LoRa

The authors themselves write that individual human gait characteristics are not yet perfectly identifiable by this method.

Moreover:

1.Analyse the text it is suggested to improve its readability. A big problem in this regard is the authors’ use of very long sentences that are subordinately complex. In same parts, the text is written with frequent use of words usually used in a different context or not used at all in the presented area of knowledge. This clarity problem is most visible (perceptible) in the Abstract and Introduction sections.

2.In the final part of the work in Discussion (Gait Recognition Based on Wireless Signal), a comparison of the authors’ proposed method with other already published methods is presented. For a complete representation of the research area, it is worth adding other methods based on inertial sensors (mentioned in first part of the work) that are not mentioned, such as:

Rong Zhu and Zhaoying Zhou, “A real-time articulated human motion tracking using tri-axis inertial/magnetic sensors package,” in IEEE Transactions on Neural Systems and Rehabilitation Engineering, vol. 12, no. 2, pp. 295-302, June 2004, doi: 10.1109/TNSRE.2004.827825, 

and an algorithm combining discrete Fourier transform (DFT) and continuous wavelet transform (CWT), for example:

Glowinski, S., Blazejewski, A., Krzyzynski, T. (2017). Human Gait Feature Detection Using Inertial Sensors Wavelets. In: González-Vargas, J., Ibáñez, J., Contreras-Vidal, J., van der Kooij, H., Pons, J. (eds) Wearable Robotics: Challenges and Trends. Biosystems & Biorobotics, vol 16. Springer, Cham, https://doi.org/10.1007/978-3-319-46532-6_65,

and also: 

Tadano, S.; Takeda, R.; Miyagawa, H. Three Dimensional Gait Analysis Using Wearable Acceleration and Gyro Sensors Based on Quaternion Calculations. Sensors 2013, 13, 9321-9343. https://doi.org/10.3390/s130709321.

In that work, quaternion calculations are a mathematical tool used to represent rotations in three-dimensional space. They are used in the paper to calculate the orientation of the sensors attached to the human body.

Analyse the text it is suggested to improve its readability. A big problem in this regard is the authors’ use of very long sentences that are subordinately complex. In same parts, the text is written with frequent use of words usually used in a different context or not used at all in the presented area of knowledge. This clarity problem is most visible (perceptible) in the Abstract and Introduction sections.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

This paper presents a gait recognition for coal mine using Long Range Radio signal. The writing of the paper is sound and the adoption of LoRa is new. The paper can be improved in the following ways:

- Related Works: Include a literature review of the existing works on gait recognition for coal mine. 

- Experimental Results: Although using LoRa is new, however, presenting the comparison results with the existing works is essential to prove that it is reliable to use LoRa for gait recognition.

-Missing ref 8 in the References.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The present study introduces a gait recognition system that employs LoRa technology to achieve identity recognition in a single target scenario. The proposed system analyzes the impact of various targets on wireless signals to accomplish this objective. The deployment process of the system is straightforward, and its operation is user-friendly and pragmatic. In order to tackle the challenges mentioned above, an approach is taken wherein the signal variation patterns resulting from distinct targets are compared and subjected to analysis. The findings indicate that different targets, despite having comparable height and weight, elicit specific signal fluctuation patterns due to varying walking gaits. This establishes an experimental basis for LoRa-based gait identification. This work is exciting and can be helpful for many applications. After reading the manuscript, here are my comments:

1. Please clarify the novelty of this work. How the approach proposed by the authors differs from similar approaches using signals on gait recognition. There are many papers published that process the signals for gait recognition algorithm research. What is your difference from the previous works and innovation? Clearly state the novelty of this study at the end of the introduction section.
2. Motivation for the study must be given in the introduction section. What is the knowledge gap bridged by this study? More is needed for motivation, including references. Please clarify this issue.
3. In data preprocessing, How do authors process noise reduction of the received signals? Please clarify.
4. How does the size of the sample interval affect the recognition accuracy? Why 5 of the sample interval is better than 4?
5. The confusion matrices from the experiments show low recognition accuracy. How do authors convince the chosen methodology (gait recognition system by LoRa technology) that it is the proper technology To solve the problems? Please clarify.

I consider that the authors need to address these issues before their paper can be published in the journal.

The quality of the English language is fine. Just minor editing of the English language is required.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

The article deals with the somewhat novel approach of employing a new approach to gait recognition using LoRa signals, taking into account the challenging conditions found in underground coal mines. The article's content is a satisfactory elaborate applicate method and presents implementational results emphasizing result distinction.

Have no further objection.

Certain parts of the language structure need improvement, which are not of significant importance.