ICACT20230135 Question.1
Questioner: phamdinhlam@kyonggi.ac.kr 2023-02-20 오전 10:57:31 |
ICACT20230135 Answer.1
Answer by Auhor yjk1425@163.com 2023-02-20 오전 10:57:31
Chrome  Click!! |
| Why does the CNN method was used in this study, not the other methods? What are the applications of this study?
|
We use CNN because the image recognition algorithm can perform fast and accurate target recognition, so we convert the signal to the time-frequency domain for recognition. This study is mainly applied to the jamming recognition of the fuze signal to lay the foundation for the subsequent jamming signal removal. |
ICACT20230135 Question.2
Questioner: hbw017@mail.sdu.edu.cn 2023-02-20 오전 11:01:20 |
ICACT20230135 Answer.2
Answer by Auhor yjk1425@163.com 2023-02-20 오전 11:01:20
Chrome Click!! |
| This paper proposed a TFA-CNN model. What advantages does this model have compared with traditional DNN, such as complexity, model parameters, etc |
Compared with the traditional DNN, CNN has better ability to process images, and from the simulation results of the article, the recognition accuracy of CNN is much higher than that of DNN. of course, this also brings higher model complexity, but after we weigh it, we think it is acceptable. |
ICACT20230135 Question.3
Questioner: myy98@mail.sdu.edu.cn 2023-02-20 오전 11:04:31 |
ICACT20230135 Answer.3
Answer by Auhor yjk1425@163.com 2023-02-20 오전 11:04:31
Chrome Click!! |
| Why should we convert the signal into time-frequency image, and finally classify the image? |
Our research focuses on how to accurately recognize jamming signals, so it is seen as an image recognition problem. Therefore, it was decided to use CNN, which also employs a time-frequency analysis method to convert to images and then to recognize them. |
ICACT20230135 Question.4
Questioner: hbw017@mail.sdu.edu.cn 2023-02-20 오전 11:08:06 |
ICACT20230135 Answer.4
Answer by Auhor yjk1425@163.com 2023-02-20 오전 11:08:06
Chrome Click!! |
| The focus of your paper work is jamming signal recognition. Is this method applicable to most jamming scenarios? |
In this study we considered five different signals and combined them two by two to form new jamming signals, which can include most of the jamming signals. In the future, we will study and recognize some new jamming signals. |
ICACT20230135 Question.5
Questioner: franklinlu888@outlook.com 2023-02-21 오후 3:42:32 |
ICACT20230135 Answer.5
Answer by Auhor yjk1425@163.com 2023-02-21 오후 3:42:32
Chrome Click!! |
| A very nice presentation! What is the characteristic of the fuze jamming signal? Are these characteristics can aid the recognition? |
Firstly, the power of the fuze jamming signals will be higher than some traditional communication jamming signals. Secondly, the fuze jamming signal is very close to the fuze signal in frequency domain, like some advanced DRFM-based jamming signals, which are real fuze signals in nature. Therefore, in order to recognize quickly and accurately, we need at least two dimensions to obtain the characteristics of the jamming signals. In the time-frequency domain, we can clearly see the difference between these jamming signals and fuzing signals due to the high power of the jamming signals, and the difference can also be seen from the whole time-frequency diagram if they are close in power. |
ICACT20230135 Question.6
Questioner: sanjoy.basak@rma.ac.be 2023-02-24 오전 11:40:20 |
ICACT20230135 Answer.6
Answer by Auhor yjk1425@163.com 2023-02-24 오전 11:40:20
Chrome Click!! |
| Very nice presentation! How did you test the low SINR conditions? Did you add noise in the simulation domain? Have you tested the performance in a real-world low SINR scenario? |
In the simulation, we consider the effect of noise and we set SNR=10dB. only need to increase the power of the jamming signal to simulate at low SINR, but we have not tested its performance in the real jamming situation. |
