Compact Amplitude-Only Direction Finding Based on a Deep Neural Network with a Single-Patch Multi-Beam Antenna
Round 1
Reviewer 1 Report
The paper needs still a minor revision. In particular, English and References are weak. For the language, we suggest a review made by a native English speaker. For References, I invite the authors to cite recent results in fractal-wavelet analysis of antennas. Thus, I suggest adding the following references (or other ones of the same scientific weight, in accordance to the current MDPI policy).
1. A Discussion on the Significance of Geometry in Determining the Resonant Behavior of Fractal and Other Non-Euclidean Wire Antennas. IEEE Antennas Propag. Mag. 2003, 45, 9-28.
2. Fractional-Wavelet Analysis of Positive definite Distributions and Wavelets on D'(C), in Engineering Mathematics II, Silvestrov, Rancic (Eds.), Springer, pp. 337-353,2016.
3. Hyperspectral image classification using wavelet transform-based smooth ordering, Int. J. Wavelets Multiresolut. Inf. Process, 17(6), Article Number: 1950050,2019.
4. Harmonic Sierpinski Gasket and Applications, Entropy, 20(9), 714, 2018.
5. A Framework of Adaptive Multiscale Wavelet Decomposition for Signals on Undirected Graphs, IEEE Transactions on Signal Processing, Volume: 67, Issue: 7,Pages: 1696-1711, 2019.
6. Primality, Fractality and Image Analysis, Entropy, 21(3), 304, 2019.
7. Self-Similarity and the Geometric Requirements for Frequency Independence in Antennae. Fractals 19
99, 7, 79-84.
English very difficult to understand/incomprehensible
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 2 Report
In this work, the authors present an approach to increase the amount of training data for DNN-based direction finding by using a single antenna with multi-beam capability. They present the simulated design of the antenna and the modifications that must be made to reduce beam width and maintain a frequency of 2.4 GHz. Next, they show how this is fabricated and compare experimental measurements to simulated predictions. Finally, they use the designed antenna with their DNN setup to show a field test with a performance estimation probability of 97.7%, a 3.5% increase from a previously published single antenna setup. Overall this work is interesting, well done, and presented with clarity. I did not find any issues with any sections or have recommendations for changes. This paper has my recommendation for publication as it is.
Author Response
Thank you very much for your detailed review of my manuscript and for your good evaluation.
Author Response File: Author Response.pdf
Reviewer 3 Report
This paper presents a compact direction finding system based on a deep neural network (DNN) with a single-patch multi-beam antenna. It is an interesting research; however, I still have some questions that I would like to clarify.
- In the abstract, it would be clearer if specific numbers were provided to demonstrate the development or improvement over the previous work.
- The electric field distribution with different port excitation is presented in Table 1 without color levels. How do the strengths of each electric field differ or are they equal?
- Why is the length of the parasitic element set to 19 mm?
- Why was DNN chosen for this research?
- How accurate is the approach presented in this work compared to the previous approach, and what percentage of improvement in accuracy does it achieve?
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
Thank you very much.