Advertisement

Deep Learning for Wireless Communications

  • Tugba ErpekEmail author
  • Timothy J. O’Shea
  • Yalin E. Sagduyu
  • Yi Shi
  • T. Charles Clancy
Chapter
Part of the Studies in Computational Intelligence book series (SCI, volume 867)

Abstract

Existing communication systems exhibit inherent limitations in translating theory to practice when handling the complexity of optimization for emerging wireless applications with high degrees of freedom. Deep learning has a strong potential to overcome this challenge via data-driven solutions and improve the performance of wireless systems in utilizing limited spectrum resources. In this chapter, we first describe how deep learning is used to design an end-to-end communication system using autoencoders. This flexible design effectively captures channel impairments and optimizes transmitter and receiver operations jointly in single-antenna, multiple-antenna, and multiuser communications. Next, we present the benefits of deep learning in spectrum situation awareness ranging from channel modeling and estimation to signal detection and classification tasks. Deep learning improves the performance when the model-based methods fail. Finally, we discuss how deep learning applies to wireless communication security. In this context, adversarial machine learning provides novel means to launch and defend against wireless attacks. These applications demonstrate the power of deep learning in providing novel means to design, optimize, adapt, and secure wireless communications.

Keywords

Deep learning Wireless systems Physical layer End-to-end communication Signal detection and classification Wireless security 

References

  1. 1.
    Haykin, S.: Cognitive radio: brain-empowered wireless communications. IEEE J. Sel. Areas Commun. 23(2), 201–220 (2005)CrossRefGoogle Scholar
  2. 2.
    Clancy, C., Stuntebeck, H.J., O’Shea, T.: Applications of machine learning to cognitive radio networks. IEEE Trans. Wirel. Commun. 14(4), 47–52 (2007)CrossRefGoogle Scholar
  3. 3.
    O’Shea, T.J., Corgan, J., Charles Clancy, T.: Convolutional radio modulation recognition networks. Int. Conf. Eng. Appl. Neural Netw. (2016)Google Scholar
  4. 4.
    Shannon, C.E.: A mathematical theory of communication. Bell Syst. Tech. J. 27, 379–423, 623–656 (1948)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Goldsmith, A.: Joint source/channel coding for wireless channels. IEEE Veh. Technol. Conf. (VTC) (1995)Google Scholar
  6. 6.
    Sagduyu, Y.E., Ephremides, A.: Cross-layer optimization of MAC and network coding in wireless queueing tandem networks. IEEE Trans. Inf. Theory 54(2), 554–571 (2008)MathSciNetCrossRefGoogle Scholar
  7. 7.
    Samuel, N., Diskin, T., Wiesel, A.: Deep MIMO detection. In: IEEE International Workshop on Signal Processing Advances in Wireless Communications (2017)Google Scholar
  8. 8.
    Wang, T., Wen, C., Wang, H., Gao, F., Jiang, T., Jin, S.: Deep learning for wireless physical layer: opportunities and challenges. China Commun. 14(11), 92–111 (2017)CrossRefGoogle Scholar
  9. 9.
    He, H., Wen, C.-K., Jin, S., Li, G.Y.: A model-driven deep learning network for MIMO detection (2018). arXiv preprint arXiv:1809.09336
  10. 10.
    Veeravalli, V.V., Annapureddy, V.S.: Gaussian interference networks: sum capacity in the low interference regime and new outer bounds on the capacity region. IEEE Trans. Inf. Theory 55(7), 3032–3050 (2009)MathSciNetCrossRefGoogle Scholar
  11. 11.
    Han, T.S., Kobayashi, K.: A new achievable rate region for the interference channel. IEEE Trans. Inf. Theory 27(1), 49–60 (1981)MathSciNetCrossRefGoogle Scholar
  12. 12.
    Carleial, A.B.: A case where interference does not reduce capacity. IEEE Trans. Inf. Theory 21(5), 569–570 (1975)MathSciNetCrossRefGoogle Scholar
  13. 13.
    Erpek, T., Ulukus, S., Sagduyu, Y.E.: Interference regime enforcing rate maximization for non-orthogonal multiple access (NOMA). In: IEEE International Conference on Computing, Networking and Communications (ICNC) (2019)Google Scholar
  14. 14.
    O’Shea, T.J., Hoydis, J.: An introduction to deep learning for the physical layer. IEEE Trans. Cogn. Commun. Netw. 3(4), 563–575 (2017)CrossRefGoogle Scholar
  15. 15.
    Dörner, S., Cammerer, S., Hoydis, J., Brink, S.: Deep learning based communication over the air. IEEE J. Sel. Top. Signal Process. 12(1), 132–143 (2018)CrossRefGoogle Scholar
  16. 16.
    O’Shea, T.J., Erpek, T., Charles Clancy, T.: Physical layer deep learning of encodings for the MIMO fading channel. In: IEEE Annual Allerton Conference on Communication, Control, and Computing (Allerton) (2017)Google Scholar
  17. 17.
    Erpek, T., O’Shea, T.J., Charles Clancy, T.: Learning a physical layer scheme for the MIMO interference channel. In: IEEE International Conference on Communications (ICC) (2018)Google Scholar
  18. 18.
    Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT Press (2016)Google Scholar
  19. 19.
    Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization (2014). arXiv preprint arXiv:1412.6980
  20. 20.
    Aoudia, F.A., Hoydis, J.: End-to-end learning of communications systems without a channel model (2018). arXiv preprint arXiv:1804.02276
  21. 21.
    Alamouti, S.M.: A simple transmit diversity technique for wireless communications. IEEE J. Sel. Areas Commun. 16(8), 1451–1458 (1998)CrossRefGoogle Scholar
  22. 22.
    Tarokh, V., Seshadri, N., Calderbank, A.R.: Space-time codes for high data rate wireless communication: performance criterion and code construction. IEEE Trans. Inf. Theory 44(2), 744–765 (1998)MathSciNetCrossRefGoogle Scholar
  23. 23.
    Telatar, E.: Capacity of multi-antenna Gaussian channels. Eur. Trans. Telecommun. 10(6), 585–595 (1999)MathSciNetCrossRefGoogle Scholar
  24. 24.
    Yu, W., Rhee, W., Boyd, S., Cioffi, J.M.: Iterative water-filling for Gaussian vector multiple access channels. IEEE Trans. Inf. Theory 50(1), 145–151 (2004)MathSciNetCrossRefGoogle Scholar
  25. 25.
    Tan, X., Xu, W., Be’ery, Y., Zhang, Z., You, X., Zhang, C.: Improving massive MIMO belief propagation detector with deep neural network (2018). arxiv preprint arXiv:1804.01002
  26. 26.
    Liu, L., Oestges, C., Poutanen, J., Haneda, K., Vainikainen, P., Quitin, F., Tufvesson, F., Doncker, P.D.: The COST 2100 MIMO channel model. IEEE Trans. Wirel. Commun. 19(6), 92–99 (2012)CrossRefGoogle Scholar
  27. 27.
    Ioffe, S., Szegedy, C.: Batch normalization: accelerating deep network training by reducing internal covariate shift (2015). arXiv preprint arXiv:1502.03167
  28. 28.
    Hassibi, B., Hochwald, B.M.: How much training is needed in multiple-antenna wireless links? IEEE Trans. Inf. Theory 49(4), 951–963 (2003)CrossRefGoogle Scholar
  29. 29.
    Erpek, T., Sagduyu, Y.E., Shi, Y., Ponnaluri, S.: Rate optimization with distributed network coordination of multiuser MIMO communications. In: IEEE Vehicular Technology Conference (VTC) (2018)Google Scholar
  30. 30.
    Erpek, T., Sagduyu, Y.E., Shi, Y., Ponnaluri, S.: Network control and rate optimization for multiuser MIMO communications. Ad Hoc Netw. 85, 92–102 (2019)CrossRefGoogle Scholar
  31. 31.
    Chollet, F.: Keras. https://github.com/fchollet/keras (2015)
  32. 32.
    Abadi, M., et al.: TensorFlow: large-scale machine learning on heterogeneous systems. https://tensorflow.org
  33. 33.
    Federal Communications Commission: Notice of proposed rule making and order: facilitating opportunities for flexible, efficient, and reliable spectrum use employing cognitive radio technologies, ET Docket No., 03-108 (2005)Google Scholar
  34. 34.
    O’Shea, T.J., Roy, T., West, N., Hilburn, B.C.: Physical layer communications system design over-the-air using adversarial networks (2018). arXiv preprint arXiv:1803.03145
  35. 35.
    Ye, H., Li, G.Y., Juang, B.-H.F., Sivanesan, K.: Channel agnostic end-to-end learning based communication systems with conditional GAN. In: IEEE Global Communications Conference (Globecom) Workshops (2018)Google Scholar
  36. 36.
    O’Shea, T.J., Roy, T., West, N.: Approximating the void: learning stochastic channel models from observation with variational generative adversarial networks (2018). arXiv preprint arXiv:1805.06350
  37. 37.
    Grathwohl, W., Choi, D., Wu, Y., Roeder, G., Duvenaud, D.: Backpropagation through the void: optimizing control variates for black-box gradient estimation (2017). arXiv preprint arXiv:1711.00123
  38. 38.
    Raj, V., Kalyani, S.: Backpropagating through the air: deep learning at physical layer without channel models. IEEE Commun. Lett. 22(11), 2278–2281 (2018)CrossRefGoogle Scholar
  39. 39.
    Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., Bengio, Y.: Generative adversarial nets. In: Advances in Neural Information Processing Systems (NeurIPS) (2014)Google Scholar
  40. 40.
    Ettus, M.: Universal software radio peripheral (2009)Google Scholar
  41. 41.
    Gulrajani, I., Ahmed, F., Arjovsky, M., Dumoulin, V., Courville, A.: Improved training of Wasserstein GANs. In: Advances in Neural Information Processing Systems (NeurIPS) (2017)Google Scholar
  42. 42.
    O’Shea, T.J., Pemula, L., Batra, D., Charles Clancy, T.: Radio transformer networks: attention models for learning to synchronize in wireless systems (2018). arXiv preprint arXiv:1805.06350
  43. 43.
    O’Shea, T.J., Roy, T., Charles Clancy, T.: Learning robust general radio signal detection using computer vision methods. In: Asilomar Conference on Signals, Systems, and Computers (2017)Google Scholar
  44. 44.
    Akyildiz, I.F., Lee, W.-Y., Vuran, M.C., Mohanty, S.: A survey on spectrum management in cognitive radio networks. IEEE Commun. Mag. 46(4), 40–48 (2008)CrossRefGoogle Scholar
  45. 45.
    O’Shea, T.J., Roy, T., Erpek, T.: Spectral detection and localization of radio events with convolutional neural features. In: European Signal Processing Conference (2017)Google Scholar
  46. 46.
    Selvaraju, R.R., Das, A., Vedantam, R., Cogswell, M., Parikh, D., Batra, D.: Grad-CAM: why did you say that? (2016) arXiv preprint arXiv:1611.07450
  47. 47.
    Lin, M., Chen, Q., Yan, S.: Network in network (2013). arXiv preprint arXiv:1312.4400
  48. 48.
    Lee, H., Battle, A., Raina, R., Ng, A.Y.: Efficient sparse coding algorithms. In: Advances in Neural Information Processing Systems (NeurIPS) (2006)Google Scholar
  49. 49.
    Zeiler, M.D., Krishnan, D., Taylor, G.W., Fergus, R.: Deconvolutional networks. In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2010)Google Scholar
  50. 50.
    O’Shea, T.J., Roy, T., Charles Clancy, T.: Over-the-air deep learning based radio signal classification. IEEE J. Sel. Top. Signal Process. 12(1), 168–179 (2018)CrossRefGoogle Scholar
  51. 51.
    Shrivastava, A., Pfister, T., Tuzel, O., Susskind, J., Wang, W., Webb, R.: Learning from simulated and unsupervised images through adversarial training. In: Conference on Computer Vision and Pattern Recognition (CVPR) (2017)Google Scholar
  52. 52.
    Shi, Y., Sagduyu, Y.E., Davaslioglu, K., Li, J.: Generative adversarial networks for black-box API attacks with limited training data. In: IEEE Symposium on Signal Processing and Information Technology (2018)Google Scholar
  53. 53.
    Shi, Y., Sagduyu, Y.E., Davaslioglu, K., Levy, R.: Vulnerability detection and analysis in adversarial deep learning. Guide to Vulnerability Analysis for Computer Networks and Systems—An Artificial Intelligence Approach. Computer Communications and Networks. Springer, Cham (2018)Google Scholar
  54. 54.
    Mirza, M., Osindero, S.: Conditional generative adversarial nets (2014). arXiv preprint arXiv:1411.1784
  55. 55.
    Davaslioglu, K., Sagduyu, Y.E.: Generative adversarial learning for spectrum sensing. In: IEEE International Conference on Communication (ICC) (2018)Google Scholar
  56. 56.
    Tang, B., Tu, Y., Zhang, Z., Lin, Y.: Digital signal modulation classification with data augmentation using generative adversarial nets in cognitive radio networks. IEEE Access. 6, 15713–15722 (2018)CrossRefGoogle Scholar
  57. 57.
    Erpek, T., Sagduyu, Y.E., Shi, Y.: Deep learning for launching and mitigating wireless jamming attacks. IEEE Trans. Cogn. Commun. Netw. 5(1), 2–14 (2019)CrossRefGoogle Scholar
  58. 58.
    Shi, Y., Sagduyu, Y.E., Erpek, T., Davaslioglu, K., Lu, Z., Li, J.: Adversarial deep learning for cognitive radio security: jamming attack and defense strategies. In: IEEE ICC 2018 Workshop—Promises and Challenges of Machine Learning in Communications Networks (2018)Google Scholar
  59. 59.
    Hughes, B., Bothe, S., Farooq, H., Imran, A.: Generative adversarial learning for machine learning empowered self organizing 5G networks. In: IEEE International Conference on Computing, Networking and Communications (ICNC). http://bsonlab.com/images/Publications/Conferences/C18-4.pdf (2019)
  60. 60.
    Donahue, J., Krahenbuhl, P., Darrell, T.: Adversarial feature learning (2016). arXiv preprint arXiv:1605.09782
  61. 61.
    Shi, Y., Davaslioglu, K., Sagduyu, Y.E.: Generative adversarial network for wireless signal spoofing. In: ACM WiSec Workshop on Wireless Security and Machine Learning (WiseML) (2019)Google Scholar
  62. 62.
    Sagduyu, Y.E., Berry, R., Ephremides, A.: Jamming games in wireless networks with incomplete information. IEEE Commun. Mag. 49(8), 112–118 (2011)CrossRefGoogle Scholar
  63. 63.
    Sagduyu, Y.E., Ephremides, A.: A game-theoretic analysis of denial of service attacks in wireless random access. J. Wirel. Netw. 15(5), 651–666 (2009)CrossRefGoogle Scholar
  64. 64.
    Sagduyu, Y.E., Berry, R., Ephremides, A.: Jamming games for power controlled medium access with dynamic traffic. In: IEEE International Symposium on Information Theory (ISIT) (2010)Google Scholar
  65. 65.
    Sagduyu, Y.E., Berry, R., Ephremides, A.: Wireless jamming attacks under dynamic traffic uncertainty. In: IEEE International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WIOPT) (2010)Google Scholar
  66. 66.
    Sagduyu, Y.E., Berry, R., Ephremides, A.: MAC games for distributed wireless network security with incomplete information of selfish and malicious user types. In: IEEE International Conference on Game Theory for Networks (GameNets) (2009)Google Scholar
  67. 67.
    Sagduyu, Y.E.: Securing cognitive radio networks with dynamic trust against spectrum sensing data falsification. In: IEEE Military Communications Conference (MILCOM) (2014)Google Scholar
  68. 68.
    Lu, Z., Sagduyu, Y.E., Li, J.: Securing the backpressure algorithm for wireless networks. IEEE Trans. Mobile Comput. 16(4), 1136–1148 (2017)CrossRefGoogle Scholar
  69. 69.
    Lu, Z., Sagduyu, Y.E., Li, J.: Queuing the trust: secure backpressure algorithm against insider threats in wireless networks. In: IEEE Conference on Computer Communications (INFOCOM) (2015)Google Scholar
  70. 70.
    Vorobeychik, Y., Kantarcioglu, M.: Adversarial Machine Learning. Morgan & Claypool (2018)Google Scholar
  71. 71.
    Shi, Y., Sagduyu, Y.E., Grushin, A.: How to steal a machine learning classifier with deep learning. In: IEEE Symposium on Technologies for Homeland Security (2017)Google Scholar
  72. 72.
    Shi, Y., Sagduyu, Y.E.: Evasion and causative attacks with adversarial deep learning. In: IEEE Military Communications Conference (MILCOM) (2017)Google Scholar
  73. 73.
    Pi, L., Lu, Z., Sagduyu, Y.E., Chen, S.: Defending active learning against adversarial inputs in automated document classification. In: IEEE Global Conference on Signal and Information Processing (GlobalSIP) Symposium on Compressed Sensing, Deep Learning (2016)Google Scholar
  74. 74.
    Shi, Y., Sagduyu, Y.E., Davaslioglu, K., Li, J.: Active deep learning attacks under strict rate limitations for online API calls. In: IEEE Symposium Technologies for Homeland Security (2018)Google Scholar
  75. 75.
    Shi, Y., Erpek, T., Sagduyu, Y.E., Li, J.: Spectrum data poisoning with adversarial deep learning. In: IEEE Military Communications Conference (MILCOM) (2018)Google Scholar
  76. 76.
    Microsoft Cognitive Toolkit (CNTK), https://docs.microsoft.com/en-us/cognitive-toolkit
  77. 77.
    Sagduyu, Y.E., Shi, Y., Erpek, T.: IoT network security from the perspective of adversarial deep learning. In: IEEE International Conference on Sensing, Communication and Networking (SECON) Workshop on Machine Learning for Communication and Networking in IoT (2019)Google Scholar
  78. 78.
    Sadeghi, M., Larsson, E.G.: Adversarial attacks on deep-learning based radio signal classification. IEEE Wirel. Commun. Lett. 8(1), 213–216 (2019)CrossRefGoogle Scholar
  79. 79.
    Flowers, B., Buehrer, R.M., Headley, W.C.: Evaluating adversarial evasion attacks in the context of wireless communications (2019). arXiv preprint arXiv:1903.01563
  80. 80.
    Hameed, M.Z., Gyorgy, A., Gunduz, D.: Communication without interception: defense against deep-learning-based modulation detection (2019). arXiv preprint arXiv:1902.10674
  81. 81.
    Kokalj-Filipovic, S., Miller, R.: Adversarial examples in RF deep learning: detection of the attack and its physical robustness (2019). arXiv preprint arXiv:1902.06044
  82. 82.
    Han, G., Xiao, L., Poor, H.V.: Two-dimensional anti-jamming communication based on deep reinforcement learning. In: IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (2017)Google Scholar
  83. 83.
    Wu, Z., Zhao, Y., Yin, Z., Luo, H.: Jamming signals classification using convolutional neural network. In: IEEE Symposium on Signal Processing and Information Technology (ISSPIT) (2017)Google Scholar
  84. 84.
    Topal, O.A., Gecgel, S., Eksioglu, E.M., Kurt, G.: Identification of smart jammers: learning based approaches using wavelet representation (2019). arXiv preprint arXiv:1901.09424
  85. 85.
    Ferdowsi, A., Saad, W.: Deep learning for signal authentication and security in massive internet of things systems (2018). arXiv preprint arXiv:1803.00916

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Tugba Erpek
    • 1
    Email author
  • Timothy J. O’Shea
    • 1
    • 2
  • Yalin E. Sagduyu
    • 3
  • Yi Shi
    • 3
    • 4
  • T. Charles Clancy
    • 1
  1. 1.Virginia TechArlingtonUSA
  2. 2.DeepSig, Inc.ArlingtonUSA
  3. 3.Intelligent Automation, Inc.RockvilleUSA
  4. 4.Virginia TechBlacksburgUSA

Personalised recommendations