Smartphone-Based Unconstrained Step Detection Fusing a Variable Sliding Window and an Adaptive Threshold

Round 1

Reviewer 1 Report

This work is easy to follow and has clear motivation. However, the proposed method to improve the step detection is not clear enough to me. Here are my suggestions:

Minor issues:

1. Ln 124: error in grammar

2. Ln 168: what is the unit of window size 15? sample? time? Different phones have different output data rates of IMU. You shall provide this information and how do you make it consistent? All the discussion and unit in "data" point is based on which data rate?

3. Ln 198: the analysis of FFT also depends on the number of sample (window size), here you need to provide the information,. 

4. Ln 233: Step 5? the eliminated peaks are used to count steps?

5. Ln 239 to 242: it is not clear.

6. Ln 252: You said that the minimum peaks detected after sliding window filter are used as adaptive threshold but in figure 7, the black dots are not minimum, they are maximum. why? 

7. It is not clear to me why you use FIR filters AND sliding window filters. Is that mean you filter the signal twice?

8. Ln 304: It is not clear about the statement " starting points moved forward two data points". What is the initial window size? If it is 256 data points, and when you said move 2 data points, is it mean you SLIDE the current window 2 data points? If so, the endpoints also moved two data points. In your current sentence, it looks like you remove two beginning points and shorten the current size of the window (because you did not say the endpoint will be moved forward too). I suggest you draw a figure to make it clear.

9. Ln 329. Table 4.5?

10. Ln 400: the sliding mean filter is applied after FIR or directly on the original overall acceleration?

11.reference 2 and number 7 are the same

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Strengths:

 (+) The problem is important.

 (+) The experiments are convincing.

Weaknesses:

 (-) There are English issues.

 (-) References are inadequate.

 (-) The introduction must be improved.

 (-) The related work section must be enhanced.

 (-) Some improvements are needed in the description of the method.

==== ENGLISH ==== 

The writing style of this paper is not good enough. Authors should spend some time improving it so that the paper can be read more smoothly.

The paper has several typos. Authors need to proofread the paper to eliminate all of them.

==== FIGURES ==== 

The text of some figure(s) is too small. Authors should make sure that the text can be read if printed on paper.

Some figure(s) are blur. Authors should either use a higher resolution figure(s) or redo them as vector graphics.

==== REFERENCES ==== 

The literature review is incomplete. Several relevant references are missing.

The paper should be updated to include more recent references, preferably from the last 2 or 3 years. Smartphones for Step Detection in Smart Cities are interrelated and equally important, hence we would like to see more information related to smart cities in your study added to the related work. Here is a suggestion to follow and refer to D Sobnath, IU Rehman "Smart cities to improve mobility and quality of life of the visually impaired". And for some cases related to patients who could benefit from the developed app can be referred to it in “Sustainable virtual reality patient rehabilitation systems with IoT sensors using virtual smart” Sustainability 13 (9), 4716

The reference format is inconsistent. Please check the format carefully and ensure it is consistent for all references.

==== INTRODUCTION ==== 

The authors should add more references in the introduction to support the claims.

The authors need to better explain the context of this research, including why the research problem is important.

The introduction should clearly explain the key limitations of prior work that are relevant to this paper. Where it is also important to include this solution as part of some technical aspects to understand its viability in the industry which can be accomplished by utilising advanced security solutions. You can refer to the following study for more details: “A machine-learning scraping tool for data fusion in the analysis of sentiments about pandemics for supporting business decisions with human-centric AI explanations SA Kumar, MM Nasralla, I García-Magariño, H Kumar PeerJ Computer Science 7, e713, and “Area and energy-efficient shift and accumulator unit for object detection in IoT applications A Hazarika, S Poddar, MM Nasralla, H Rahaman Alexandria Engineering Journal 61 (1), 795-809”

Contributions should be highlighted more. It should be made clear what is novel and how it addresses the limitations of prior work. 

==== RELATED WORK ==== 

The related work section is not well organized. Authors must try to categorize the papers and present them in a logical way.

The authors should explain clearly what  the differences are between the prior work and the solution presented in this paper.

==== PROBLEM DEFINITION ==== 

The authors should add a clear and detailed problem definition. 

==== METHOD ==== 

The authors should first give an overview of their solution before explaining the details. 

A novel solution is presented but it is important to better explain the design decisions (e.g. why the solution is designed like that)

==== CONCLUSION ==== 

Some text must be added to discuss future work or research opportunities.

The conclusion should be written in a narrative format rather than bullet points.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The article needs a more complete conclusion to demonstrate the advantages of the solution

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Thanks for addressing the comments.