A Robust Target Tracking Method for Crowded Indoor Environments Using mmWave Radar
Round 1
Reviewer 1 Report
In this study, the authors present a novel approach to group tracking, encompassing a sequence of methods such as signal pre-processing, Adaptive Extended Kalman Filtering (AEKF), group association, target initialization, false target suppression, track management, and track re-association. To validate the feasibility and effectiveness of the proposed methodology, multiple experiments were conducted and discussed. The manuscript raises several concerns that need to be addressed:
1. It is unclear whether the authors have compared their proposed algorithm with any existing methods to demonstrate its effectiveness.
2. As a target tracking method, the manuscript lacks a discussion on the accuracy of the tracking process.
3. The performance of the tracking algorithm appears to be closely related to human behavior. However, the manuscript does not provide a detailed description of how human behavior, such as speed and interpersonal spacing, is accurately represented in the experiments.
4. Although the experiments demonstrate the algorithm's ability to remove false targets, perform track re-association, and estimate expansion, there seems to be a lack of corresponding evaluation metrics.
Author Response
Please see the attachment.
Author Response File: 
Author Response.doc
Reviewer 2 Report
I really like the work you did here. It is relatively rare to see theoretical considerations augmented by the experimental verification. I particularly liked the variety of the experiments you set up to determine the accuracy of your algorithm (Figures 15 and 17).
I have no outstanding technical issues with the material presented. The paper is nicely written. I would like to suggest some small changes:
Less common acronyms need spelling out first time you use them:
· Line 45: First appearance of GNN and JPDA
· Line 143 (Figure 3): first appearance of OS-CFAR and 2D MVDR. OS-CFAR could be spelled out in text, line 176.
· Line 185: CA-CFAR
No other issues. Thank you for the great work.
Author Response
Please see the attachment.
Author Response File: Author Response.doc
Reviewer 3 Report
This paper proposed alpha-extended-Kalman filter for constantly estimating the target expansion and the number of reflection points. And also utilized a density-based spatial clustering approach. Through the results, the authors have tried to establish continuous and steady tracking with high accuracy.
There are some observations as follows,
1. The paper is well presented but the novelty can be more prominent if it includes some comparative analysis with the existing work at least in terms of accuracy.
2. It will be better if authors include a pseudo-code for Alpha-Extended-Kalman Filter.
3. Please label the Z-axis in Fig. 8 and 9.
4. Please make sure that all the abbreviations are defined first.
5. Table 2 needs a more detailed discussion as it is giving different impressions under different conditions.
6. The conclusion section can be further improved by including more quantitative analysis. Also, include the future scope of work.
Author Response
Please see the attachment.
Author Response File: Author Response.doc
Round 2
Reviewer 3 Report
It is improved. Hence the paper is recommended for the publication.
