在電磁波的傳送下，每位用戶都能輕易使用無線通訊信號與遠方的其他使用者進行溝通，但各個用戶的通訊信號也會互相影響，更值得注意的是通信訊號的攔截與惡意攻擊也比有線通訊更加容易，因此在無線通訊帶來方便的情況下，如何保護使用者的安全隱私成為了一個巨大的挑戰，傳統如確認MAC與IP位置是常見的資訊安全保護方法，但是若有刻意模仿的情形下也不能保證訊息的安全，因此射頻指紋特徵(RF Fingerprint)成為資安防護的新方法。
射頻指紋特徵主要透過不同發射機的硬體缺陷來進行分類，因不同發射機的缺陷各有不同並且這些缺陷是非常難進行模仿與複製，所以利用射頻指紋特徵能在實體層中對發射機進行確認並對用戶進行授權，能大幅減少被模仿的可能。但是指紋特徵不僅在發射端受缺陷影響，同時也會在接收端的硬體設備受到影響，因此在某接收端所得的訓練模型很難移轉到不同接收端上使用，致使實際應用上會有普及的挑戰與困難。
本論文利用域對抗神經網路(Domain-Adversarial Neural Network) 的機器學習方法，對不同接收端所收集的資料進行域的映射，使得目標域(Target Domain)的映射能接近甚至等同於源域(Source Domain)的映射，如此一來就能利用源資料集(Source Data)所訓練的分類模型對目標資料集(Target Data)進行分類，換句話說可以利用其中一台接收端資料的訓練模型在另一台接收端上進行分類，進而達到接收端轉移的效果。
本論文利用五台不同的軟體無線電(Universal Software Radio Peripheral, USRP)共10個接收端，進行數據測量並據以進行域對抗神經網路學習。實驗結果表示，此方法能提升接收端轉移的準確率，進而有更好的分類結果。

Under the transmission of electromagnetic waves, each user can easily use wireless communication signals to communicate with other users far away. However, the communication signals of each user will also affect each other. More noteworthy is that the signal interception and malicious attacks of wireless communication are also easier than wired communication. Therefore, how to protect the security and privacy of users has become a huge challenge in the convenient wireless communications. Traditional information security protection methods such as confirming MAC and IP location are common, which cannot be guaranteed in the case of deliberate imitation. Therefore, RF Fingerprint has become a new method for security protection.
The radio frequency fingerprints are mainly classified by the unique hardware defects of different transmitters. Because the defects of different transmitters are very difficult to imitate and copy, the physical layer radio frequency fingerprints can be used to authenticate the transmitters and to authorize the corresponding users.
As the transmitter RF fingerprints are affected by both the transmitter end and the receiver end, the training model obtained by a receiver end is hard to transfer to another receiver end. Thus, the problem impedes the pervasive applications of this technology.
This work uses the Domain-Adversarial Neural Network (DANN) machine learning method to map the data collected by different receivers. The mapping of the target domain is close to or even equal to the source domain, so that the classification model trained by the source data set can be used to classify the target data set. In other words, the training model of one receiver end can be transferred and used for another receiver.
This work uses five different Universal Software Radio Peripheral, that is 10 receivers in total, for signal measurement. The measurement data are used for DANN method.
The experimental results show that this method can improve the classification accuracy when the training model is transferred to a different receiver.

[1] W. Wang, Z. Sun, S. Piao, B. Zhu, and K. Ren, "Wireless Physical-Layer Identification: Modeling and Validation," IEEE Transactions on Information Forensics and Security, vol. 11, no. 9, pp. 2091-2106, 2016.
[2] N. Soltanieh, Y. Norouzi, Y. Yang, and N. C. Karmakar, "A Review of Radio Frequency Fingerprinting Techniques," IEEE Journal of Radio Frequency Identification, pp. 1-1, 2020.
[3] L. Peng, A. Hu, J. Zhang, Y. Jiang, J. Yu, and Y. Yan, "Design of a Hybrid RF Fingerprint Extraction and Device Classification Scheme," IEEE Internet of Things Journal, vol. 6, no. 1, pp. 349-360, 2019.
[4] L. Chia-Ling, "Impacts of I/Q imbalance on QPSK-OFDM-QAM detection," IEEE Transactions on Consumer Electronics, vol. 44, no. 3, pp. 984-989, 1998.
[5] A. Tarighat, R. Bagheri, and A. H. Sayed, "Compensation schemes and performance analysis of IQ imbalances in OFDM receivers," IEEE Transactions on Signal Processing, vol. 53, no. 8, pp. 3257-3268, 2005.
[6] S. Raschka, V. J. S.-L. Mirjalili, and T. S. e. ed, "Python Machine Learning: Machine Learning and Deep Learning with Python," 2017.
[7] D. Harris and S. Harris, Digital design and computer architecture. San Francisco, CA, USA: Morgan Kaufmann, 2010.
[8] R. Bellman, "Dynamic programming," Science, vol. 153, no. 3731, pp. 34-37, 1966.
[9] J. C. Ang, A. Mirzal, H. Haron, and H. N. A. Hamed, "Supervised, Unsupervised, and Semi-Supervised Feature Selection: A Review on Gene Selection," IEEE/ACM Transactions on Computational Biology and Bioinformatics, vol. 13, no. 5, pp. 971-989, 2016.
[10] G. Chandrashekar and F. Sahin, "A survey on feature selection methods," Computers & Electrical Engineering, vol. 40, no. 1, pp. 16-28, 2014.
[11] I. Guyon and E. Elisseeff, "An introduction to variable and feature selection," JMLR, vol. 3, no. Mar, pp. 1157-1182, 2003.
[12] K. S. Balagani and V. V. Phoha, "On the Feature Selection Criterion Based on an Approximation of Multidimensional Mutual Information," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 32, no. 7, pp. 1342-1343, 2010.
[13] P. Hanchuan, L. Fuhui, and C. Ding, "Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 27, no. 8, pp. 1226-1238, 2005.
[14] N. Soltanieh, Y. Norouzi, Y. Yang, and N. C. Karmakar, "A Review of Radio Frequency Fingerprinting Techniques," IEEE Journal of Radio Frequency Identification, vol. 4, no. 3, pp. 222-233, 2020.
[15] G. Baldini and G. Steri, "A Survey of Techniques for the Identification of Mobile Phones Using the Physical Fingerprints of the Built-In Components," IEEE Communications Surveys & Tutorials, vol. 19, no. 3, pp. 1761-1789, 2017.
[16] M. Taha, R. Atallah, O. Dwiek, and F. Bata, "Crowd Estimation Based on RSSI Measurements Using kNN Classification," in 2020 3rd International Conference on Intelligent Autonomous Systems (ICoIAS), 2020, pp. 67-70.
[17] M. Ezuma, F. Erden, C. K. Anjinappa, O. Ozdemir, and I. Guvenc, "Micro-UAV Detection and Classification from RF Fingerprints Using Machine Learning Techniques," in 2019 IEEE Aerospace Conference, 2019, pp. 1-13.
[18] I. O. Kennedy, P. Scanlon, F. J. Mullany, M. M. Buddhikot, K. E. Nolan, and T. W. Rondeau, "Radio Transmitter Fingerprinting: A Steady State Frequency Domain Approach," in 2008 IEEE 68th Vehicular Technology Conference, 2008, pp. 1-5.
[19] S. U. Rehman, K. Sowerby, C. Coghill, and W. Holmes, "The analysis of RF fingerprinting for low-end wireless receivers with application to IEEE 802.11a," in 2012 International Conference on Selected Topics in Mobile and Wireless Networking, 2012, pp. 24-29.
[20] S. Raschka and V. Mirjalili, Python Machine Learning: Machine Learning and Deep Learning with Python, scikit-learn, and TensorFlow 2. Packt Publishing Ltd, 2019.
[21] S. Deng, Z. Huang, and X. Wang, "A novel specific emitter identification method based on radio frequency fingerprints," in 2017 2nd IEEE International Conference on Computational Intelligence and Applications (ICCIA), 2017, pp. 368-371.
[22] S. Chen, F. Xie, Y. Chen, H. Song, and H. Wen, "Identification of wireless transceiver devices using radio frequency (RF) fingerprinting based on STFT analysis to enhance authentication security," in 2017 IEEE 5th International Symposium on Electromagnetic Compatibility (EMC-Beijing), 2017, pp. 1-5.
[23] B. Kroon, S. Bergin, I. O. Kennedy, and G. O. M. Zamora, "Steady state RF fingerprinting for identity verification: one class classifier versus customized ensemble," in Irish Conference on Artificial Intelligence and Cognitive Science, 2009, pp. 198-206: Springer.
[24] S. V. Radhakrishnan, A. S. Uluagac, and R. Beyah, "GTID: A Technique for Physical Device and Device Type Fingerprinting," IEEE Transactions on Dependable and Secure Computing, vol. 12, no. 5, pp. 519-532, 2015.
[25] A. S. Uluagac, S. V. Radhakrishnan, C. Corbett, A. Baca, and R. Beyah, "A passive technique for fingerprinting wireless devices with Wired-side Observations," in 2013 IEEE Conference on Communications and Network Security (CNS), 2013, pp. 305-313.
[26] L. Zong, C. Xu, and H. Yuan, "A RF Fingerprint Recognition Method Based on Deeply Convolutional Neural Network," in 2020 IEEE 5th Information Technology and Mechatronics Engineering Conference (ITOEC), 2020, pp. 1778-1781.
[27] Y. Ganin and V. Lempitsky, "Unsupervised Domain Adaptation by Backpropagation," presented at the Proceedings of the 32nd International Conference on Machine Learning, Proceedings of Machine Learning Research, 2015.
[28] Y. Ganin et al., "Domain-adversarial training of neural networks," J. Mach. Learn. Res, vol. 17, no. 1, pp. 2096-2030, 2016.
[29] "IEEE Standard for Telecommunications and Information Exchange Between Systems - LAN/MAN Specific Requirements - Part 11: Wireless Medium Access Control (MAC) and physical layer (PHY) specifications: High Speed Physical Layer in the 5 GHz band," IEEE Std 802.11a-1999, pp. 1-102, 1999.
[30] R. Battiti, "Using mutual information for selecting features in supervised neural net learning," IEEE Transactions on Neural Networks, vol. 5, no. 4, pp. 537-550, 1994.
[31] N. Kwak and C. Chong-Ho, "Input feature selection for classification problems," IEEE Transactions on Neural Networks, vol. 13, no. 1, pp. 143-159, 2002.