AI for 5G: research directions and paradigms


Wireless communication technologies such as fifth generation mobile networks (5G) will not only provide an increase of 1000 times in Internet traffic in the next decade but will also offer the underlying technologies to entire industries to support Internet of things (IOT) technologies. Compared to existing mobile communication techniques, 5G has more varied applications and its corresponding system design is more complicated. The resurgence of artificial intelligence (AI) techniques offers an alternative option that is possibly superior to traditional ideas and performance. Typical and potential research directions related to the promising contributions that can be achieved through AI must be identified, evaluated, and investigated. To this end, this study provides an overview that first combs through several promising research directions in AI for 5G technologies based on an understanding of the key technologies in 5G. In addition, the study focuses on providing design paradigms including 5G network optimization, optimal resource allocation, 5G physical layer unified acceleration, end-to-end physical layer joint optimization, and so on.

This is a preview of subscription content, access via your institution.


  1. 1

    You X H, Pan Z W, Gao X Q, et al. The 5G mobile communication: the development trends and its emerging key techniques (in Chinese). Sci Sin Inform, 2014, 44: 551–563

    Article  Google Scholar 

  2. 2

    Li L M, Wang D M, Niu X K, et al. mmWave communications for 5G: implementation challenges and advances. Sci China Inf Sci, 2018, 61: 021301

    Article  Google Scholar 

  3. 3

    Wang C X, Wu S B, Bai L, et al. Recent advances and future challenges for massive MIMO channel measurements and models. Sci China Inf Sci, 2016, 59: 021301

    Google Scholar 

  4. 4

    Zhang J H, Tang P, Tian L, et al. 6–100 GHz research progress and challenges from a channel perspective for fifth generation (5G) and future wireless communication. Sci China Inf Sci, 2017, 60: 080301

    Article  Google Scholar 

  5. 5

    Tao X F, Han Y, Xu X D, et al. Recent advances and future challenges for mobile network virtualization. Sci China Inf Sci, 2017, 60: 040301

    Article  Google Scholar 

  6. 6

    3GPP. Way forward on the overall 5G-NR eMBB. Workplan RP-170741. 2017. RAN/TSG RAN/TSGR 75/Docs/

  7. 7

    3GPP. Study on new radio access technology: radio access architecture and interfaces (release 14). TR38.801, v14.0. 2017.

  8. 8

    ITU-R. Minimum requirements related to technical performance for IMT2020 radio interface(s). Report ITU-RM.2410-0. 2017.

  9. 9

    3GPP. LTE Enhancements and 5G Normative Work. Release-15. 2018.

  10. 10

    You X H, Wang D M, Sheng B, et al. Cooperative distributed antenna systems for mobile communications. IEEE Wirel Commun, 2010, 17: 35–43

    Article  Google Scholar 

  11. 11

    Yang W J, Wang M, Zhang J J, et al. Narrowband wireless access for low-power massive internet of things: a bandwidth perspective. IEEE Wirel Commun, 2017, 24: 138–145

    Article  Google Scholar 

  12. 12

    ITU-T. LS/o on the results of the 1st meeting of the ITU-T focus group on machine learning for future networks including 5G (FG ML5G). FG ML5G-0-004. 2018.

  13. 13

    3GPP. 5G system network data analytics services stage 3. TS 29.520 (CT3). 2018.

  14. 14

    Whitley D. A genetic algorithm tutorial. Stat Comput, 1994, 4: 65–85

    Article  Google Scholar 

  15. 15

    Schmidhuber J. Deep learning in neural networks: an overview. Neural Netw, 2015, 61: 85–117

    Article  Google Scholar 

  16. 16

    You X H, Chen G A, Cheng S X. Dynamic learning rate optimization of the backpropagation algorithm. IEEE Trans Neural Netw, 1995, 6: 669–677

    Article  Google Scholar 

  17. 17

    You X H. Can backpropagation error surface not have local minima. IEEE Trans Neural Netw, 1992, 3: 1019–1021

    Article  Google Scholar 

  18. 18

    Yu X H, Chen G A. Efficient backpropagation learning using optimal learning rate and momentum. Neural Netw, 1997, 10: 517–527

    Article  Google Scholar 

  19. 19

    Kaelbling L P, Littman M L, Moore A W. Reinforcement learning: a survey. J Artif Intell Res, 1996, 4: 237–285

    Article  Google Scholar 

  20. 20

    Watkins C J C H, Dayan P. Q-learning. Mach Learn, 1992, 8: 279–292

    MATH  Google Scholar 

  21. 21

    Finn C, Abbeel P, Levine S. Model-agnostic meta-learning for fast adaptation of deep networks. 2017. ArXiv: 1703.03400

    Google Scholar 

  22. 22

    Wu J X, Gao B B, Wei X S, et al. Resource-constrained deep learning: challenges and practices. Sci Sin Inform, 2018, 48: 501–510

    Article  Google Scholar 

  23. 23

    Zhou Z H. Machine learning: recent progress in China and beyond. China Sci Rev, 2018, 5: 20

    Google Scholar 

  24. 24

    Zhong Y X. Artificial intelligence: concept, approach and opportunity. Chin Sci Bull, 2017, 62: 2473

    Article  Google Scholar 

  25. 25

    Gatherer A. Machine learning modems: how ML will change how we specify and design next generation communication systems. IEEE ComSoc Tech News, 2018.

    Google Scholar 

  26. 26

    Yang C, Xu W H, Zhang Z C, et al. A channel-blind detection for SCMA based on image processing techniques. In: Proceedings of IEEE International Symposium on Circuits and Systems (ISCAS), 2018. 1–5

    Google Scholar 

  27. 27

    Zhang C, XuW H. Neural networks: efficient implementations and applications. In: Proceedings of IEEE International Conference on ASIC (ASICON), 2017. 1029–1032

    Google Scholar 

  28. 28

    Xu W H, You X H, Zhang C. Efficient deep convolutional neural networks accelerator without multiplication and retraining. In: Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2018. 1–5

    Google Scholar 

  29. 29

    Xu W H, Wang Z F, You X H, et al. Efficient fast convolution architectures for convolutional neural network. In: Proceedings of IEEE International Conference on ASIC (ASICON), 2017. 904–907

    Google Scholar 

  30. 30

    Xu W H, Wu Z Z, Ueng Y L, et al. Improved polar decoder based on deep learning. In: Proceedings of IEEE International Workshop on Signal Processing Systems (SiPS), 2017. 1–6

    Google Scholar 

  31. 31

    Xu W H, Zhong Z W, Be’ery Y, et al. Joint neural network equalizer and decoder. In: Proceedings of IEEE International Symposium on Wireless Communication Systems (ISWCS), 2018. 1–6

    Google Scholar 

  32. 32

    Xu W H, Be’ery Y, You X H, et al. Polar decoding on sparse graphs with deep learning. In: Proceedings of Asilomar Conference on Signals, Systems, and Computers (Asilomar), 2018. 1–6

    Google Scholar 

  33. 33

    Xu W H, You X H, Zhang C. Using Fermat number transform to accelerate convolutional neural network. In: Proceedings of IEEE International Conference on ASIC (ASICON), 2017. 1033–1036

    Google Scholar 

  34. 34

    Gao X Q, Jiang B, Li X, et al. Statistical eigenmode transmission over jointly correlated MIMO channels. IEEE Trans Inform Theor, 2009, 55: 3735–3750

    Article  MathSciNet  MATH  Google Scholar 

  35. 35

    Wang D M, Zhang Y, Wei H, et al. An overview of transmission theory and techniques of large-scale antenna systems for 5G wireless communications. Sci China Inf Sci, 2016, 59: 081301

    Article  Google Scholar 

  36. 36

    Gesbert D, Hanly S, Huang H, et al. Multi-cell MIMO cooperative networks: a new look at interference. IEEE J Sel Areas Commun, 2010, 28: 1380–1408

    Article  Google Scholar 

  37. 37

    Jing S S, Yu A L, Liang X, et al. Uniform belief propagation processor for massive MIMO detection and GF (2n) LDPC decoding. In: Proceedings of IEEE International Conference on ASIC (ASICON), 2017. 961–964

    Google Scholar 

  38. 38

    Gandhi V S, Maheswaran B. A cross layer design for performance enhancements in LTE-A system. In: Proceedings of IEEE International Conference on Wireless Communications, Signal Processing and Networking (WiSPNET), 2016. 905–909

    Google Scholar 

  39. 39

    Kuen J, Kong X F, Wang G, et al. DelugeNets: deep networks with efficient and flexible cross-layer information inflows. In: Proceedings of IEEE International Conference on Computer Vision Workshop (ICCVW), 2017. 958–966

    Google Scholar 

  40. 40

    Farsad N, Rao M, Goldsmith A. Deep learning for joint source-channel coding of text. 2018. ArXiv: 1802.06832

    Google Scholar 

  41. 41

    Xu X W, Ding Y K, Hu S X, et al. Scaling for edge inference of deep neural networks. Nat Electron, 2018, 1: 216–222

    Article  Google Scholar 

  42. 42

    Wang X F, Li X H, Leung V C M. Artificial intelligence-based techniques for emerging heterogeneous network: state of the arts, opportunities, and challenges. IEEE Access, 2015, 3: 1379–1391

    Article  Google Scholar 

  43. 43

    Klaine P V, Imran M A, Onireti O, et al. A survey of machine learning techniques applied to self-organizing cellular networks. IEEE Commun Surv Tut, 2017, 19: 2392–2431

    Article  Google Scholar 

  44. 44

    Pèrez-Romero J, Sallent O, Ferrús R, et al. Knowledge-based 5G radio access network planning and optimization. In: Proceedings of IEEE International Symposium on Wireless Communication Systems (ISWCS), 2016. 359–365

    Google Scholar 

  45. 45

    Gómez-Andrades A, Munoz P, Serrano I, et al. Automatic root cause analysis for LTE networks based on unsupervised techniques. IEEE Trans Veh Technol, 2016, 65: 2369–2386

    Article  Google Scholar 

  46. 46

    Wang J H, Guan W, Huang Y M, et al. Distributed optimization of hierarchical small cell networks: a GNEP framework. IEEE J Sel Areas Commun, 2017, 35: 249–264

    Article  Google Scholar 

  47. 47

    Bogale T E, Wang X, Le L B. Machine intelligence techniques for next-generation context-aware wireless networks. 2018. ArXiv: 1801.04223

    Google Scholar 

  48. 48

    Li R, Zhao Z, Zhou X, et al. Intelligent 5G: when cellular networks meet artificial intelligence. IEEE Wirel Commun, 2017, 24: 175–183

    Article  Google Scholar 

  49. 49

    Zhao Z, Li R, Sun Q, et al. Deep reinforcement learning for network slicing. 2018. ArXiv: 1805.06591

    Google Scholar 

  50. 50

    Ren Y R, Zhang C, Liu X, et al. Efficient early termination schemes for belief-propagation decoding of polar codes. In: Proceedings of IEEE International Conference on ASIC (ASICON), 2015. 1–4

    Google Scholar 

  51. 51

    Fossorier M P C, Mihaljevic M, Imai H. Reduced complexity iterative decoding of low-density parity check codes based on belief propagation. IEEE Trans Commun, 1999, 47: 673–680

    Article  Google Scholar 

  52. 52

    Yang J M, Song W Q, Zhang S Q, et al. Low-complexity belief propagation detection for correlated large-scale MIMO systems. J Sign Process Syst, 2018, 90: 585–599

    Article  Google Scholar 

  53. 53

    Liu L, Yuen C, Guan Y L, et al. Gaussian message passing iterative detection for MIMO-NOMA systems with massive access. In: Proceedings of IEEE Global Communications Conference (GLOBECOM), 2016. 1–6

    Google Scholar 

  54. 54

    Yang J M, Zhang C, Zhou H Y, et al. Pipelined belief propagation polar decoders. In: Proceedings of IEEE International Symposium on Circuits and Systems (ISCAS), 2016. 413–416

    Google Scholar 

  55. 55

    Tan X S, Xu W H, Be’ery Y, et al. Improving massive MIMO belief propagation detector with deep neural network. 2018. ArXiv: 1804.01002

    Google Scholar 

  56. 56

    Liang F, Shen C, Wu F. An iterative BP-CNN architecture for channel decoding. IEEE J Sel Top Signal Process, 2018, 12: 144–159

    Article  Google Scholar 

  57. 57

    Lv X Z, Wei P, Xiao X C. Automatic identification of digital modulation signals using high order cumulants. Electronic Warfare, 2004, 6: 1

    Google Scholar 

  58. 58

    Wang T Q, Wen C K, Wang H Q, et al. Deep learning for wireless physical layer: opportunities and challenges. China Commun, 2017, 14: 92–111

    Article  Google Scholar 

  59. 59

    O’Shea T, Hoydis J. An introduction to deep learning for the physical layer. IEEE Trans Cogn Commun Netw, 2017, 3: 563–575

    Article  Google Scholar 

  60. 60

    O’Shea T J, Erpek T, Clancy T C. Deep learning based MIMO communications. 2017. ArXiv: 1707.07980

    Google Scholar 

Download references


This work was supported by National Natural Science Foundation of China (Grant Nos. 61501116, 61521061).

Author information



Corresponding author

Correspondence to Chuan Zhang.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

You, X., Zhang, C., Tan, X. et al. AI for 5G: research directions and paradigms. Sci. China Inf. Sci. 62, 21301 (2019).

Download citation


  • 5G mobile communication
  • AI techniques
  • network optimization
  • resource allocation
  • unified acceleration
  • end-to-end joint optimization