Multi-Task Learning with Dynamic Splitting for Open-Set Wireless Signal Recognition

doi:10.12142/ZTECOM.2022S1007

ZTE Communications ›› 2022, Vol. 20 ›› Issue (S1): 44-56.DOI: 10.12142/ZTECOM.2022S1007

• Research Paper • Previous Articles Next Articles

Multi-Task Learning with Dynamic Splitting for Open-Set Wireless Signal Recognition

XU Yujie¹, ZHAO Qingchen¹, XU Xiaodong¹(), QIN Xiaowei¹, CHEN Jianqiang²

^1.University of Science and Technology of China, Hefei 230026, China
^2.ZTE Corporation, Shenzhen 518057, China

Received:2021-07-05 Online:2022-01-25 Published:2022-03-01
About author:XU Yujie received the B. Eng. and M. Eng. degree in electronic and information engineering from University of Science and Technology of China (USTC), China in 2017 and 2021. His research interest is deep learning for wireless signal identification.|ZHAO Qingchen received the B.Eng. degree in electronic information engineering from Anhui University of Finance and Economics, China in 2018. He is currently pursuing his M. Eng. degree in Department of Electronic Engineering and Information Science, University of Science and Technology of China (USTC), China. His research interest is smart transmission.|XU Xiaodong (xdxu@ustc.edu.cn) received the B.Eng. and Ph.D. degrees in electronic and information engineering from University of Science and Technology of China (USTC) in 2000 and 2007, respectively. From 2000 to 2001, he served as a research assistant at the R&D center, Konka Telecommunications Technology. Since 2007, he has been a faculty member with the Department of Electronic Engineering and Information Science, USTC. He is currently working with the CAS Key Laboratory of Wireless-Optical Communications, USTC. His research interests include the areas of wireless communications, signal processing, wireless artificial intelligence and information-theoretic security.|QIN Xiaowei received the B.S. and Ph.D. degrees from the Department of Electrical Engineering and Information Science, University of Science and Technology of China (USTC), China in 2000 and 2008, respectively. Since 2014, he has been a member of staff in Key Laboratory of Wireless-Optical Communications of Chinese Academy of Sciences at USTC. His research interests include optimization theory, service modeling in future heterogeneous networks, and wireless artificial intelligence in mobile communication networks.|CHEN Jianqiang received the M. Eng. degree from Nanjing University of Technology, China in electromagnetic field and microwave technology. At present, he works at the intelligent home terminal product line of ZTE Corporation and has many years of experience in the communication industry. His research direction is Wi-Fi key technologies and their applications, in which he has more than 10 patents.
Supported by:
the Natural Science Foundation of Anhui Province(2008085MF213);ZTE Industry-University-Institute Cooperation Funds(20190822003)

Abstract

Abstract:

Open-set recognition (OSR) is a realistic problem in wireless signal recognition, which means that during the inference phase there may appear unknown classes not seen in the training phase. The method of intra-class splitting (ICS) that splits samples of known classes to imitate unknown classes has achieved great performance. However, this approach relies too much on the predefined splitting ratio and may face huge performance degradation in new environment. In this paper, we train a multi-task learning (MTL) network based on the characteristics of wireless signals to improve the performance in new scenes. Besides, we provide a dynamic method to decide the splitting ratio per class to get more precise outer samples. To be specific, we make perturbations to the sample from the center of one class toward its adversarial direction and the change point of confidence scores during this process is used as the splitting threshold. We conduct several experiments on one wireless signal dataset collected at 2.4 GHz ISM band by LimeSDR and one open modulation recognition dataset, and the analytical results demonstrate the effectiveness of the proposed method.

Key words: open-set recognition, dynamic method, adversarial direction, multi-task learning, wireless signal

XU Yujie, ZHAO Qingchen, XU Xiaodong, QIN Xiaowei, CHEN Jianqiang. Multi-Task Learning with Dynamic Splitting for Open-Set Wireless Signal Recognition[J]. ZTE Communications, 2022, 20(S1): 44-56.

Figures/Tables 11

Figure 1 An example of known known classes (KKCs) and UUCs using t-SNE (t-distributed stochastic neighbor embedding)

Figure 2 Multi-task learning (MTL) network architecture of the proposed method

Figure 3 An example of confidence score approximate curve

Figure 4 Pictures of main transmitters used in WS dataset collection

Table 1 WS dataset parameters

Wi-Fi, Bluetooth, cordless phone, wide-band FM,

ZigBee, microwave oven, analog video monitor,

narrow-band digital signal, wide-band OFDM

signal, game control signal

Line-of-sight (1, 3, 5, 7 m);

Non-line-of-sight (1, 3 m)

2.442 GHz

20 MHz

7 500

Figure 5 T of proposed methods with and without MTL on the radio modulation dataset (RML) dataset

Figure 6 T of proposed methods with and without MTL on the wireless signal dataset (WS) dataset

Figure 7 T of the proposed, OSISR and ICS methods under different O on the radio modulation dataset (RML) dataset

Figure 8 T of the proposed, OSISR and ICS methods under different O on the wireless signal dataset (WS) dataset

Figure 9 T of proposed methods using different data augmentation methods on the radio modulation dataset (RML)

Figure 10 T of proposed methods using different data augmentation methods on the wireless signal dataset (WS)

References 34

1	RAYANCHU S, PATRO A, BANERJEE S. Airshark: detecting non-WiFi RF devices using commodity WiFi hardware [C]//ACM SIGCOMM conference on Internet measurement conference. ACM, 2011: 137–154. DOI: 10.1145/2068816.2068830 DOI URL
2	LI X F, DONG F W, ZHANG S, et al. A survey on deep learning techniques in wireless signal recognition [J]. Wireless communications and mobile computing, 2019, 2019: 5629572. DOI: 10.1155/2019/5629572 DOI URL
3	DOBRE O A, HAMEED F. Likelihood-based algorithms for linear digital modulation classification in fading channels [C]//Canadian Conference on Electrical and Computer Engineering. IEEE, 2006: 1347–1350. DOI: 10.1109/CCECE.2006.277525 DOI URL
4	CHAVALI V G, SILVA C R C MDA. Maximum-likelihood classification of digital amplitude-phase modulated signals in flat fading non-Gaussian channels [J]. IEEE transactions on communications, 2011, 59(8): 2051–2056. DOI: 10.1109/TCOMM.2011.051711.100184 DOI URL
5	PALICOT J, ROLAND C. A new concept for wireless reconfigurable receivers [J]. IEEE communications magazine, 2003, 41(7): 124–132. DOI: 10.1109/MCOM.2003.1215649 DOI URL
6	KIM K, AKBAR I A, BAE K K, et al. Cyclostationary approaches to signal detection and classification in cognitive radio [C]//IEEE International Symposium on New Frontiers in Dynamic Spectrum Access Networks. IEEE, 2007: 212–215. DOI: 10.1109/DYSPAN.2007.35 DOI URL
7	KIM Y, AN S, SO J. Identifying signal source using channel state information in wireless LANs [C]//International Conference on Information Networking (ICOIN). IEEE, 2018: 616–621. DOI: 10.1109/ICOIN.2018.8343192 DOI URL
8	YANG Z S, WANG Y X, ZHANG L J, et al. Indoor interference classification based on WiFi channel state information [C]//Proc. International Conference on Security, Privacy and Anonymity in Computation, Communication and Storage. Springer, 2018: 136–145. DOI: 10.1007/978-3-030-05345-1_11 DOI URL
9	KULIN M, KAZAZ T, MOERMAN I, et al. End-to-end learning from spectrum data: A deep learning approach for wireless signal identification in spectrum monitoring applications [J]. IEEE access, 2018, 6: 18484–18501. DOI: 10.1109/ACCESS.2018.2818794 DOI URL
10	RIYAZ S, SANKHE K, IOANNIDIS S, et al. Deep learning convolutional neural networks for radio identification [J]. IEEE communications magazine, 2018, 56(9): 146–152. DOI: 10.1109/MCOM.2018.1800153 DOI URL
11	NAYLOR A R. Known knowns, known unknowns and unknown unknowns: a 2010 update on carotid artery disease [J]. The surgeon, 2010, 8(2): 79–86. DOI: 10.1016/j.surge.2010.01.006 DOI URL
12	O’SHEA T J, ROY T, CLANCY T C. Over-the-air deep learning based radio signal classification [J]. IEEE journal of selected topics in signal processing, 2018, 12(1): 168–179. DOI: 10.1109/JSTSP.2018.2797022 DOI URL
13	GENG C X, HUANG S J, CHEN S C. Recent advances in open set recognition: A survey [J]. IEEE transactions on pattern analysis and machine intelligence, 2021, 43(10): 3614–3631. DOI: 10.1109/TPAMI.2020.2981604 DOI URL
14	MENDES JÚNIOR P R, SOUZA R M, DE O WERNECK R, et al. Nearest neighbors distance ratio open-set classifier [J]. Machine learning, 2017, 106(3): 359–386. DOI: 10.1007/s10994-016-5610-8 DOI URL
15	BENDALE A, BOULT T E. Towards open set deep networks [C]//IEEE Conference on Computer Vision and Pattern Recognition. IEEE, 2016: 1563–1572. DOI: 10.1109/CVPR.2016.173 DOI URL
16	SCHEIRER W J, ROCHA A, MICHEALS R J, et al. Meta-recognition: The theory and practice of recognition score analysis [J]. IEEE transactions on pattern analysis and machine intelligence, 2011, 33(8): 1689–1695. DOI: 10.1109/TPAMI.2011.54 DOI URL
17	YOSHIHASHI R, SHAO W, KAWAKAMI R, et al. Classification-reconstruction learning for open-set recognition [C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, 2019: 4011–4020. DOI: 10.1109/CVPR.2019.00414 DOI URL
18	NEAL L, OLSON M, FERN X, et al. Open set learning with counterfactual images [C]//Proc. European Conference on Computer Vision (ECCV). Springer, 2018: 613–628
19	JO I, KIM J, KANG H, et al. Open set recognition by regularising classifier with fake data generated by generative adversarial networks [C]//IEEE International Conference on Acoustics, Speech and Signal Processing. IEEE, 2018: 2686–2690. DOI: 10.1109/ICASSP.2018.8461700 DOI URL
20	YU Y, QU W Y, LI N, et al. Open-category classification by adversarial sample generation [EB/OL]. (2017-06-17)[2020-10-15].
21	CRAWSHAW M. Multi-task learning with deep neural networks: A survey [EB/OL]. (2020-09-10)[2021-10-15].
22	PERERA P, MORARIU V I, JAIN R, et al. Generative-discriminative feature representations for open-set recognition [C]//IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, 2020: 11811–11820. DOI: 10.1109/CVPR42600.2020.01183 DOI URL
23	OZA P, PATEL V M. Deep CNN-based multi-task learning for open-set recognition [EB/OL]. (2019-03-07)[2020-10-15].
24	YU Q, IKAMI D, IRIE G, et al. Multi-task curriculum framework for open-set semi-supervised learning [C]//European Conference on Computer Vision. Springer, 2020: 438–454. DOI: 10.1007/978-3-030-58610-2_26 DOI URL
25	XU Y J, QIN X W, XU X D, et al. Open-set interference signal recognition using boundary samples: A hybrid approach [C]//International Conference on Wireless Communications and Signal Processing (WCSP). IEEE, 2020: 269–274. DOI: 10.1109/WCSP49889.2020.9299700 DOI URL
26	YI S, WANG H, XUE W Q, et al. Interference source identification for IEEE 802.15.4 wireless sensor networks using deep learning [C]//29th Annual International Symposium on Personal, Indoor and Mobile Radio Communications. IEEE, 2018: 1–7. DOI: 10.1109/PIMRC.2018.8580857 DOI URL
27	CROCE D, GARLISI D, GIULIANO F, et al. Learning from errors: detecting cross-technology interference in WiFi networks [J]. IEEE transactions on cognitive communications and networking, 2018, 4(2): 347–356. DOI: 10.1109/TCCN.2018.2816068 DOI URL
28	SCHLACHTER P, LIAO Y W, YANG B. Open-set recognition using intra-class splitting [C]//27th European Signal Processing Conference (EUSIPCO). IEEE, 2019: 1–5. DOI: 10.23919/eusipco.2019.8902738 DOI URL
29	SCHLACHTER P, LIAO Y W, YANG B. Deep open set recognition using dynamic intra-class splitting [J]. SN computer science, 2020, 1(2): 1–12. DOI: 10.1007/s42979-020-0086-9 DOI URL
30	MIYATO T, MAEDA S I, KOYAMA M, et al. Virtual adversarial training: a regularization method for supervised and semi-supervised learning [J]. IEEE transactions on pattern analysis and machine intelligence, 2019, 41(8): 1979–1993. DOI: 10.1109/TPAMI.2018.2858821 DOI URL
31	KIM D, CHOI Y, KIM Y. Understanding and improving virtual adversarial training [EB/OL]. (2019-09-15)[2020-10-15].
32	SONG M K, WANG Z B, ZHANG Z F, et al. Analyzing user-level privacy attack against federated learning [J]. IEEE journal on selected areas in communications, 2020, 38(10): 2430–2444. DOI: 10.1109/JSAC.2020.3000372 DOI URL
33	ZACK G W, ROGERS W E, LATT S A. Automatic measurement of sister chromatid exchange frequency [J]. The journal of histochemistry and cytochemistry, 1977, 25(7): 741–753. DOI: 10.1177/25.7.70454 DOI URL
34	HUANG L, PAN W J, ZHANG Y, et al. Data augmentation for deep learning-based radio modulation classification [J]. IEEE access, 2019, 8: 1498–1506. DOI: 10.1109/ACCESS.2019.2960775 DOI URL

Multi-Task Learning with Dynamic Splitting for Open-Set Wireless Signal Recognition

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 11

References 34

Related Articles 0

Recommended Articles

Metrics