Grant-free DSA-NOMA with Enhanced MUD and K-repetition for B5G mMTC : 차세대 5G 대규모 기기용 통신을 위한 향상된 다중 사용자 검출 및 K-반복을 적용한 그랜트 프리 다이버시티 슬롯 알로하 비직교 다중 접속

다국어 초록 (Multilingual Abstract)

Massive Machine Type Communication (mMTC) is defined as a vital scenario in fifth-generation (5G) networks and beyond (B5G), aiming to enable reliable and efficient uplink (UL) transmissions for Internet of Things (IoT) applications in the presence of massive devices with sporadic traffic and small-size data. Combining diversity slotted ALOHA (DSA), K-repetition grant-free access (K-GFA) and non-orthogonal multiple access (NOMA) into GF-DSA-NOMA is considered promising, as the concomitant techniques and design principles have the potential to engender an efficient and reliable UL access paradigm within the context of mMTC.
The focus of this thesis is to provide feasible solutions for multi-user detection (MUD) and robust transmission in the face of collision issues within the context of GF-DSA-NOMA. Firstly, we take a step towards designing a blind inter resource block iterative interference cancellation (inter-RB-IIC) scheme for GF-DSA-NOMA system, leveraging memory mechanism of access points (APs) for MUD. Next, we investigate a K-GFA scheme for GF-DSA-NOMA system, called (QK, Q)-K-GFA, incorporating the erasure code for error correction. The goal is to achieve high reliability and low latency access in the presence of massive uncoordinated IoT devices. Lastly, we present a deep reinforcement learning (DRL) framework to dynamically manage the repetition strategy of devices while optimizing blind inter-RB-IIC parameters, aiming to maximize the access probability performance of the GF-DSA-NOMA system while minimizing memory write-read complexity caused by the proposed blind inter-RB-IIC.
We present an analysis of the complexity of the blind inter-RB-IIC mechanism. We derive closed-form analytical models for the blind inter-RB-IIC and (QK, Q)-K-GFA to calculate the access probability of devices. Extensive numerical simulation are conducted to validate the efficacy of the proposed analytical models, and gain insights into the performance of the proposed blind inter-RB-IIC and (QK, Q)-K-GFA across various key operational parameters. The effectiveness of proposed DRL-based solution is also validated by simulation results.

국문 초록 (Abstract)

대규모 기기용 통신(mMTC)은 5G 및 그 이후의 모바일 시스템에서 중요한 시나리오로 정의되며, 드문 트래픽과 소규모 데이터가 있는 환경 에서 사물인터넷(IoT) 애플리케이션을 위한 신뢰성 있고 효율적인 업링크(UL) 전송을 가능하게 하는 것을 목표로 한다. 다이버시티 슬롯 알로하 (DSA), K-반버 그랜트 프리 접속(K-GFA), 비직교 다중 접속(NOMA)를 결합한 GF-DSA-NOMA는 유망한 접근 방식으로 간주되며, 관련 기술 및 설계 원칙이 mMTC 맥락에서 효율적이고 신뢰할 수 있는 UL 액세스 패러다임을 창출할 가능성이 있다.
이 논문의 초점은 GF-DSA-NOMA 맥락에서 다중 사용자 검출(MUD) 및 충돌 문제에 직면한 강력한 전송에 대한 실현 가능한 해결책을 제공하는 것인다. 먼저, MUD를 위한 액세스 포인트(AP)의 메모리 메커니즘을 활용하여 GF-DSA-NOMA 시스템에 대한 블라인드 인터 리소스 블록 반복 간섭 제거(inter-RB-IIC) 기법을 설계하는 단계로 나아간다. 다음으로, 오류 수정용 소거 코드를 통합한 (QK, Q)-K-GFA라는 GF-DSA-NOMA 시스템에 대한 K-GFA 기법을 개발한다. 목표는 대규모 비조정 IoT 장치의 존재하에 높은 신뢰성과 낮은 지연 시간 액세스를 달성하는 것인다. 마지막으로, 제안된 블라인드 inter-RB-IIC에 의해 발생하는 메모리 쓰기-읽기 복잡성을 최소화하면서 GF-DSA-NOMA 시스템의 접근 확률 성능을 극대화하기 위해 장치의 반복 전략을 동적으로 관리하고 블라인드 inter-RB-IIC 매개변수를 최적화하는 심층 강화 학습(DRL) 프레임워크를 제시한다.
본 논문은 inter-RB-IIC 메커니즘의 복잡도에 대한 분석을 제시한다. 장치의 액세스 확률을 계산하기 위해 inter-RB-IIC 및 (QK, Q)-K-GFA에 대한 폐쇄형 해석 모델을 도출한다. 제안된 해석 모델의 효능을 검증하고 다양한 주요 운영 매개변수에 걸쳐 제안된 inter-RB-IIC 및 (QK, Q)-K-GFA의 성능에 대한 통찰을 얻기 위해 광범위한 수치 시뮬 레이션을 수행한다. 제안된 DRL 기반 솔루션의 효과도 시뮬레이션 결과로 검증된다.

목차 (Table of Contents)

• I. Introduction 1
• 1.1 Background 1
• 1.2 Motivations 5
• 1.2.1 Existing Problems 5
• 1.2.2 Proposed Grant-free Diversity Slotted ALOHA based Non-orthogonal Multiple Access (DSA-NOMA) 7
• 1.3 Contributions and Organization of the Thesis 10
• II. Related Works 13
• 2.1 K-repetition Grant-free Access (K-GFA) at User side for Grant-free Non-orthogonal Multiple Access (NOMA) 13
• 2.2 Multi-user Detection (MUD) at AP side for Grant-free NOMA 15
• 2.3 Integration of Diversity Slotted ALOHA and NOMA: DSA-NOMA 18
• III. Blind inter-RB-Iterative Interference Cancellation (IIC) 21
• 3.1 System Model 21
• 3.2 Design of Blind inter-RB-IIC Method 25
• 3.2.1 Generic Implement Model of Blind inter-RB-IIC 28
• 3.2.2 Complexity Analysis 34
• 3.3 Analytical Models 38
• 3.4 Numerical Case Study 44
• 3.5 Chapter Summary 54
• IV. (QK , Q)-K-GFA 55
• 4.1 System Model 55
• 4.2 Design of (QK , Q)-K-GFA Scheme 57
• 4.3 Analytical Models 60
• 4.4 Numerical Case Study 64
• 4.5 Chapter Summary 70
• V. Deep Reinforcement Learning (DRL)-based Optimization 72
• 5.1 Motivations 72
• 5.2 Prerequisites 74
• 5.2.1 Deep Reinforcement Learning 74
• 5.2.2 Multi-layer Perceptron (MLP) 75
• 5.3 System Model 77
• 5.4 Proposed DRL-based Optimization Framework 79
• 5.4.1 Overview of Proposed DRL Framework 79
• 5.4.2 MLP based Successful Access Probability and Complexity Predictions 84
• 5.4.3 DRL-based Optimization Mechanism 85
• 5.5 Numerical Analysis 95
• 5.6 Chapter Summary 102
• VI. Conclusion 104
• References 106
• 국문초록 112
• Appendix A 114
• Appendix B 115
• Appendix C 117

참고문헌 (Reference)

1. Backpropagation, Y. Chauvin, R. Durbin, R. Golden and, D. E. Rumelhart, Psychology Press, , 2013

2. Time-offset aloha with sic,, Y. Hu and, H. Jin, J.-B. Seo, IEEE Transactions on Mobile Computing, vol. 22, no. 11, pp. 6817–6829, 2023, , 2023

3. Physical layer procedures for data, G. NR, document 3GPP TS 38.214 v15.9.0 release 15, , 2020

4. Revisiting discrete soft actor-critic, Z. Lin, J. Li, D. Ye, W. Yang and, Q. Fu, H. Zhou, arXiv preprint arXiv:2209.10081, , 2022

5. Deep reinforcement learning: An overview, Y. Li, arXiv preprint arXiv:1701.07274, , 2017

6. On the capacity of a cellular cdma system,, K. Gilhousen, A. Viterbi, L. Weaver and, I. Jacobs, C. Wheatley, R. Padovani, vol. 40, no. 2, pp. 303–312, , 1991

7. Performance analysis of noma random access,, H. Jin, J.-B. Seo, B. C. Jung and, vol. 22, no. 11, pp. 2242–2245, , 2018

8. Optimal repetition rate for maximal coverage,, A. Al-Hourani and, B. A. Homssi, B. Manzoor, S. Kandeepan, vol. 10, no. 4, pp. 800–804, , 2021

9. Soft actor-critic algorithms and applications,, A. Zhou, S. Ha, H. Zhu, A. Gupta, P. Abbeel and, T. Haarnoja, S. Levine, J. Tan, K. Hartikainen, G. Tucker, V. Kumar, arXiv preprint arXiv:1812.05905, , 2019

10. Soft actor-critic for discrete action settings, P. Christodoulou, arXiv preprint arXiv:1910.07207, , 2019

11. 8Nr; nr and ng-ran overall description; stage 2, T. V15.4.0, 3GPP, , 2018

12. Noma-based random access with multichannel aloha,, J. Choi, vol. 35, no. 12, pp. 2736–2743, , 2017

13. Ofdma mobile broadband communications: A systems approach, R. Laroia, J. Li, X. Wu and, Cambridge University Press, , 2013

14. Grant-free non-orthogonal multiple access for iot: A survey, M. Shirvanimoghaddam and, M. B. Shahab, R. Abbas, S. J. Johnson, IEEE Communications Surveys Tutorials, vol. 22, no. 3, pp. 1805–1838, , 2020

15. Reinforcement learning for sequential decision and optimal control, S. E. Li, 2023, , 2023

16. Noma-based irregular repetition slotted aloha for satellite networks,, M. Yang, Z. Sun, X. Shao, S. Gu and, Q. Guo, vol. 23, no. 4, pp. 624–627, , 2019

17. Diversity aloha - a random access scheme for satellite communications,, G. Choudhury and, S. Rappaport, IEEE Transactions on Communications, vol. 31, no. 3, pp. 450–457, , 1983

18. Effective erasure codes for reliable computer communication protocols,, L. Rizzo, ACM SIGCOMM computer communication review, vol. 27, no. 2, pp. 24–36, , 1997

19. Noma-based crdsa with access control for next-generation iot networks,, I. N. A. Ramatryana and, S. Y. Shin, 2021 International Conference on Information and Communication Technology Convergence (ICTC), pp. 997–1001, , 2021

20. Deep reinforcement learning-based grant-free noma optimization for murllc,, Y. Liu, Y. Deng, M. Elkashlan and, H. Zhou, A. Nallanathan, vol. 71, no. 3, pp. 1475–1490, 2023, , 2023

21. System level analysis of k-repetition for uplink grant-free urllc in 5g nr, K. Pedersen, I. Z. Kovacs and, G. Berardinelli, R. Abreu, T. Jacobsen, P. Mogensen, European Wireless 2019; 25th European Wireless Conference, pp. 1–5, , 2019

22. Coded slotted aloha: A graphbased method for uncoordinated multiple access,, M. Chiani, E. Paolini, G. Liva and, IEEE Transactions on Information Theory, vol. 61, no. 12, pp. 6815–6832, , 2015

23. Contentionbased access for ultra-reliable low latency uplink transmissions,, Z. Li and, O. Tirkkonen, B. Singh, M. A. Uusitalo, vol. 7, no. 2, pp. 182–185, , 2018

24. Performance analysis of grant-free random-access noma in urll iot networks,, M. R. Amini and, M. W. Baidas, IEEE Access, vol. 9, pp. 105 974–105 988, , 2021

25. Reinforcement learning and dynamic programming using function approximators, B. De Schutter and, R. Babuska, D. Ernst, L. Busoniu, CRC Press, , 2017

26. Energy-efficient multiple access scheme with power control for mmtc networks,, D. Wu and, Z. Zhang, K. Peng, Y. Meng, 2023 IEEE 23rd International Conference on Communication Technology (ICCT), pp. 1462–1467, 2023, , 2023

27. On uplink non-orthogonal multiple access for 5g: opportunities and challenges,, C. Yan, Y. Yuan, W. Cao and, W. Li, Z. Yuan, L. Tian, vol. 14, no. 12, pp. 142–152, , 2017

28. Adaptive repetition control using terminal mobility for uplink grant-free urllc,, H. Suganuma and, S. Ozaku, F. Maehara, Y. Shimbo, 2020 IEEE 91st Vehicular Technology Conference (VTC2020- Spring), pp. 1–5, , 2020

29. Radio resource allocation and retransmission schemes for urllc over 5g networks,, S. E. Elayoubi, P. Brown, M. Deghel and, A. Galindo-Serrano, vol. 37, no. 4, pp. 896–904, , 2019

30. Adaptive transmission parameters selection algorithm for urllc traffic in uplink,, A. Shahsin, E. Khorov, A. Krasilov and, A. Belogaev, 2020 International Conference Engineering and Telecommunication (EnT), pp. 1–5, , 2020

31. Uplink vs. downlink noma in cellular networks: Challenges and research directions,, H. Tabassum, M. S. Ali, E. Hossain, D. I. Kim, M. J. Hossain and, 2017 IEEE 85th Vehicular Technology Conference (VTC Spring), pp. 1–7, , 2017

32. 5G wireless network slicing for eMBB, URLLC, and mMTC: A communication-theoretic view, K. F. Trillingsgaard, O. Simeone and, P. Popovski, G. Durisi, IEEE Access, vol. 6, pp. 55 765– 55 779, , 2018

33. Iterative collision resolution for slotted aloha with noma for heterogeneous devices,, H.-P. Cheng, S.-L. Shieh, Y.-C. Huang, Y.-P. Hsu and, vol. 69, no. 5, pp. 2948–2961, , 2021

34. Graph-based analysis and optimization of contention resolution diversity slotted aloha,, G. Liva, IEEE Transactions on Communications, vol. 59, no. 2, pp. 477–487, , 2011

35. Mobile radio communications: Second and third generation cellular and watm systems: 2nd, R. Steele and, L. Hanzo, IEEE Press- John Wiley, , 1999

36. Deepreinforcement- learning-based noma-aided slotted aloha for leo satellite iot networks,, H. Yu, Y. Gong, Z. Chen and, Z. Fei, H. Zhao, J. Wang, vol. 10, no. 20, pp. 17 772–17 784, 2023, , 2023

37. On the fundamental limits of random non-orthogonal multiple access in cellular massive iot,, M. Dohler and, S. J. Johnson, M. Condoluci, M. Shirvanimoghaddam, vol. 35, no. 10, pp. 2238–2252, , 2017

38. Deep reinforcement learning for throughput improvement of the uplink grant-free noma system,, X. Tao, N. Zhang and, H. Wu, X. Zhang, J. Zhang, vol. 7, no. 7, pp. 6369–6379, , 2020

39. Intelligent resource allocation in backscatter-noma networks: A soft actor critic framework,, M. Fayaz and, A. Nallanathan, W. Ahsan, A. Alajmi, IEEE Transactions on Vehicular Technology, vol. 72, no. 8, pp. 10 119–10 132, 2023, , 2023

40. Transmit power pool design for grant-free noma-iot networks via deep reinforcement learning,, M. Fayaz, Y. Liu and, W. Yi, A. Nallanathan, IEEE Transactions on Wireless Communications, vol. 20, no. 11, pp. 7626–7641, , 2021

41. Adaptive traffic load in irsa-noma prioritizing emergency devices for 6g enabled massive iot,, S. Y. Shin, G. B. Satrya and, I. N. A. Ramatryana, vol. 10, no. 12, pp. 2713–2717, , 2021

42. Fast q-learning for improved finite length performance of irregular repetition slotted aloha,, N. Thomos, E. Nisioti and, IEEE Transactions on Cognitive Communications and Networking, vol. 6, no. 2, pp. 844–857, , 2020

43. Deep Dyna-reinforcement learning based on random access control in LEO satellite IoT networks,, V. C. M. Leung, H. Zhang, X. Liu, K. Long, A. Nallanathan and, vol. 9, no. 16, pp. 14 818–14 828, , 2022

44. An adaptive and reliable forward error correction mechanism for real-time video delivery from uavs,, A. Zahran, B. Sarvi, C. J. Sreenan and, 2023 14th International Conference on Network of the Future (NoF), pp. 141–149, 2023, , 2023

45. Intra-slot interference cancellation for collision resolution in irregular repetition slotted aloha,, G. Araniti, S. Pfletschinger, G. Interdonato, F. Vázquez-Gallego, J. Alonso- Zarate and, 2015 IEEE International Conference on Communication Workshop (ICCW), pp. 2069–2074, , 2015

46. Deep reinforcement learning for the joint aoi and throughput optimization of the random access system,, H. Zhao, H. Yu, Z. Fei, Z. Zhang, M. Zeng and, 2022 14th International Conference on Wireless Communications and Signal Processing (WCSP), pp. 695–700, , 2022

47. Multi-agent drl approach for energy-efficient resource allocation in urllc-enabled grant-free noma systems,, D.-D. Tran, S. K. Sharma, S. Chatzinotas and, V. N. Ha, I. Woungang, vol. 4, pp. 1470–1486, 2023, , 2023

48. Grantfree access for mmtc: A performance analysis based on number of preambles, repetitions, and retransmissions,, N. M. Balasubramanya, M. Rana and, H. Grama Srinath, vol. 9, no. 16, pp. 15 169–15 183, , 2022

49. Contention resolution diversity slotted aloha (crdsa): An enhanced random access schemefor satellite access packet networks,, E. Casini, O. Del Rio Herrero, R. De Gaudenzi and, vol. 6, no. 4, pp. 1408–1419, , 2007

50. Optimization of energy efficiency for uplink murllc over multiple cells using cooperative multiagent reinforcement learning,, Q. Song, F.-C. Zheng and, J. Luo, IEEE Internet of Things Journal, vol. 11, no. 9, pp. 16 351–16 363, 2024, , 2024

51. Irregular repetition slotted aloha over rayleigh block fading channels: Bounds and threshold saturation via spatial coupling,, J. Cheng, T. Kimura and, G. Song, Y. Takahashi, IEEE Access, vol. 11, pp. 106 528–106 543, 2023, , 2023

52. Improved grant-free access for urllc via multi-tier-driven computing: Network-load learning, prediction, and resource allocation,, Q. Du and, Z. Zhao, G. K. Karagiannidis, vol. 41, no. 3, pp. 607–622, 2023, , 2023

53. Toward massive machine type communications in ultra-dense cellular iot networks: Current issues and machine learning-assisted solutions,, X. Wang, S. K. Sharma and, IEEE Communications Surveys Tutorials, vol. 22, no. 1, pp. 426–471, , 2020