Improved Model-Based Rao and Wald Test for Adaptive Range-Spread Target Detection^†

Round 1

Reviewer 1 Report

1.Line 63, what is C?

2.Equation (1) – H1: why it is n not nt ?

3.Lines 70 and 148 : what is ‘s’ following the nt ?

4.Line 76: it’s should be its

5.Line 91: transpose operation can be defined at line 68

6.Line 121: Therrien needs referencing

7.Line 134-137 “to achieve the same asymptotic performance as GLRT” needs proof.

8.Line 140: Fig . 2 I think should be Fig. 1

9.Fig. 1 requires some explanation.

10.In Fig. 1,  how are the two decision outputs used?

11.Fig. 1 – what is Secondary data? I assume this refers to the training data. If so, then the training data should also be referred to as secondary data on Line 65.

12.The following acronyms should be defined on first use: AR, CFAR (in the introduction), IAR

13.Section 2 and 3 are both titled problem formulation

14.The paper introduction can better state the contributions, e.g from the conclusion:

“This paper proposes the design of two detectors for circumstances of range-spread targets detection, in the presence of disturbance which is in possession of unknown covariance, based on two-step design”

15. The approach seems to be motivated by the small data problem (Line 51). If so, this is not clear in the paper’s purpose, and could be highlighted better. 

16. Section 4. should provide more details about the simulated data. How much data?

17. Section 5 provides results on real data for the IAR detectors only. It would be interesting to compare the IAR with the AR as well on the real data.

 

 

 

Author Response

Thank you very much for your professional comments. Please see the attachment for your specific reply. From your comments, we can see that you are a very excellent scholar and prudent reviewer.

Author Response File: Author Response.pdf

Reviewer 2 Report

34 "Fortunately, exploiting the AR property of the disturbance can effectively reduce the data set required for the estimation of unknown parameters [13]."

- (AR) acronym should be detailed / explained at its first occurrence in the text.

48   "Eventually, the conclusion can be summed up by analyzing simulation results, which demonstrates that the proposed IAR-Rao and IAR-Wald detectors have the ability to achieve a higher level of detection performance than existing ones for small data record."

- (IAR) acronym should be detailed / explained at its first occurrence in the text.

157   "That is because the traditional methods of AR detectors discard P samples. When the pulse number N becomes larger, the traditional AR detectors become very close to the IAR detectors. This phenomenon is caused by the truth that the number of the discarded pluses (P) can be ignored as N increase

- (P) acronym should be detailed / explained at its first occurrence in the text. It is true that it has been presented in the text, but at the second occurence.

Author Response

Thank you very much for your professional comments. Please see the attachment for your specific reply. From your comments, we can see that you are a very excellent scholar and prudent reviewer.

Author Response File: Author Response.pdf

Reviewer 3 Report

Nice and interesting paper on adaptive detector design in the presence of AR-modeled disturbance. Still, before publication, it is my opinion that the following major comments should be addressed by the authors:

1)    Abstract – “attain constant false alarm rate(CFAR)” -> “attain constant false alarm rate (CFAR)”

2)    Sec. I – Please avoid the use of contracted forms in a technical paper, e.g. “it’s required to set the training data…”.

3)    All the acronyms should be defined in their first use, e.g. “AR”.

4)    Please add a notation paragraph at the end of Sec. I.

5)    The following closely—related works on adaptive detection (both including one- and two-step methods, and invariance techniques) have been missed by the authors
"A unifying framework for adaptive radar detection in homogeneous plus structured interference—Part I: On the maximal invariant statistic." IEEE Transactions on Signal Processing 64.11 (2016): 2894-2906.

"Adaptive radar detection of a subspace signal embedded in subspace structured plus Gaussian interference via invariance." IEEE Transactions on Signal Processing 64.8 (2015): 2156-2167.

"On the statistical invariance for adaptive radar detection in partially homogeneous disturbance plus structured interference." IEEE Transactions on Signal Processing 65.5 (2016): 1222-1234.

6)    At the end of Sec. 3, I would like the authors to briefly discuss the computational complexity involved in the proposed Rao and Wald AR detection techniques.

7)    Most of the ROC and PD vs. SNR figure should be redrawn aiming at improved reproduction quality.

8)    Conclusion section should be enriched with a brief paragraph highlighting future directions of research. One possible option may be the derivation of two-step Gradient and Durbin tests based on the AR covariance estimation, following:

"A unifying framework for adaptive radar detection in homogeneous plus structured interference—Part II: Detectors design." IEEE Transactions on Signal Processing 64.11 (2016): 2907-2919.

Author Response

Thank you very much for your professional comments. Please see the attachment for your specific reply. From your comments, we can see that you are a very excellent scholar and prudent reviewer.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The revision has addressed my comments to my satisfaction. I recommend this article for publication. 

Reviewer 3 Report

Nice and interesting paper on adaptive detector design in the presence of AR-modeled disturbance. 

The authors have satisfactorily addressed my previous comments and modified their manuscript accordingly.