在長程演進網路之中，基地台會週期性廣播系統資訊(system information)給用戶設備。為了要改善獲取系統資訊這個過程中無線資源的使用以及造成的延遲，第五代行動通訊系統的新無線電技術將系統資訊分為最小系統資訊和其他系統資訊這兩大類，並且也規範了隨選系統資訊(on-demand system information)傳送，這是一種讓基地台根據用戶設備的需求來傳送其他系統資訊的新方法。此篇論文我們討論了隨選系統資訊傳送的細節，包含以第一訊息要求和以第三訊息要求隨選系統資訊的設計。此外，我們也同時考慮波束掃動和「Listen Before Request」這種用戶行為的影響。我們提出了總共十種傳送方式的選擇，利用它們相對應的分析模型來評估信令負擔、延遲、用戶設備的電力消耗以及基地台波束數量對結果的影響。根據分析模型和模擬的結果，可以提出對應不同系統狀況下系統資訊傳輸的最佳策略。

In LTE, the system information (SI) is periodically broadcasted to UEs. To improve the utilization of wireless resources and the latency of SI acquisition, SI is categorized into minimum SI and other SI in 5G NR network. Moreover, a new approach, on-demand SI delivery, is also specified in NR to allow gNBs transmitting other SI based on the request of UEs. In this paper, discuss the details of on-demand SI delivery, including MSG1 and MSG3 based designs. Besides, the effects of the beam sweeping and the UE behavior, "Listen Before Request", are considered. Up to 10 delivery options are introduced, and corresponding analytical models are proposed to evaluate the signaling overhead, delay, UE power consumption and the effects of the number of BS beams. On the basis of the analytical and simulation results, the policies of SI delivery for different system situations are also proposed.

摘要iii
Abstract v
1 Introduction 1
2 Related work 5
3 Distribution of on demand system information 7
3.1 SI request . . . . . . . . . . . . . . . . . . 7
3.1.1 Request mechanism . . . . . . . . . . . . . . 7
3.1.2 Listen Before Request (LBR) . . . . . . . . 10
3.2 SI delivery . . . . . .. . . . . . . . . . . . 11
3.2.1 Delivery mechanism . . . . . . . . . . . . . 11
4 Design options 15
4.0.1 PB . . . . . . . . . . . . . . . . . . . . . 15
4.0.2 OB (M1, S, L) . . . . . . . . . . . . . . . 16
4.0.3 OB (M1, S) . . . . . . . . . . . . . . . . . 16
4.0.4 OB (M1, L) . . . . . . . . . . . . . . . . . 16
4.0.5 OB (M1) . . . . . . . . . . . . . . . . . . 17
4.0.6 OB (M3, S, L) . . . . . . . . . . . . . . . 17
4.0.7 OB (M3, S) . . . . . . . . . . . . . . . . . 18
4.0.8 OB (M3, L) . . . . . . . . . . . . . . . . . 18
4.0.9 OB (M3) . . . . . . . . . . . . . . . . . . 19
4.0.10 OU . .. . . . . . . . . . . . . . . . . . . 19
5 Mathematical formulation 21
5.1 System Model . . . . . . . . . . . . . . . . . 21
5.1.1 MSG1/MSG3 based . . . . . . . . . . . . . . 21
5.1.2 Beam sweeping . . . . . . . . . . . . . . . 22
5.1.3 LBR . . . . . . . . . . . . . . . . . . . . 23
5.2 Performance Evaluation Metrics . . . . . . . . 26
5.2.1 Signaling overhead, Ωs . . . . . . . . . . . 26
5.2.2 Delay, Ωd . . . . . . . . . . . . . . . . . 28
5.2.3 UE power consumption, Ωp . . . . . . . . . . 28
5.3 Analysis of all possible cases . . . . . . . . 29
5.3.1 PB . . . . . . . . . . . . . . . . . . . . . 29
5.3.2 OB (M1, S, L) . . . . . . . . . . . . . . . 29
5.3.3 OB (M1, S) . . . . . . . . . . . . . . . . . 32
5.3.4 OB (M1, L) . . . . . . . . . . . . . . . . . 32
5.3.5 OB (M1) . . . . . . . . . . . . . . . . . . 33
5.3.6 OB (M3, S, L) . . . . . . . . . . . . . . . 34
5.3.7 OB (M3, S) . . . . . . . . . . . . . . . . . 36
5.3.8 OB (M3, L) . . . . . . . . . . . . . . . . . 37
5.3.9 OB (M3) . . . . . . . . . . . . . . . . . . 37
5.3.10 OU . . . . . . . . . . . . . . . . . . . . 38
6 Simulation Results 39
6.1 Methodology . . . . . . . . . . . . . . . . . 39
6.2 Different arrival rate of SI demands . . . . . 40
6.2.1 Signaling overhead . . . . . . . . . . . . . 40
6.2.2 Delay . . . . . . . . . . . . . . . . . . . 41
6.2.3 UE power consumption . . . . . . . . . . . . 43
6.3 Different number of BS beams . . . . . . . . . 44
6.3.1 Signaling overhead . . . . . . . . . . . . . 44
6.3.2 Delay . . . . . . . . . . . . . . . . . . . 45
6.3.3 UE power consumption . . . . . . . . . . . . 45
6.4 Utility function . . . . . . . . . . . . . . . 46
7 Conclusions 51
8 Appendix 53
8.1 Derivation of the expected average delay of the delivery options without
LBR . . . . . . . . . . . . . . . . . . . . . . . 53
8.1.1 Base case . . . . . . . . . . . . . . . . . 55
8.1.2 Inductive Hypothesis . . . . . . . . . . . . 55
8.1.3 Inductive step . . . . . . . . . . . . . . . 56
Bibliography 57

[1] 3GPP. DRX parameters in LTE. Tdoc R2071286, 3rd Generation Partnership Project (3GPP), 03 2007. Nokia Tdoc to TSNRAN WG2 #57bis meeting.
[2] 3GPP. Ondemand System Information Acquisition. Tdoc 2164948, 3rd Generation Partnership Project (3GPP), 08 2016. MediaTek Inc. Tdoc to TSNRAN2 #95 meeting.
[3] 3GPP. Ondemand SI Delivery: Signaling Aspects. Tdoc R21700011, 3rd Generation Partnership Project (3GPP), 01 2017. Samsung Tdoc to TSNRAN WG2 NR meeting.
[4] 3GPP. Nr; overall description; stage2. Technical Specification (TS) 38.300, 3rd Generation Partnership Project (3GPP), 4 2019. Version 15.5.0.
[5] 3GPP. NR; Physical channels and modulation. Technical Specification (TS) 38.211, 3rd Generation Partnership Project (3GPP), 03 2019. Version 15.5.0.
[6] 3GPP. Nr; radio resource control (rrc); protocol specification. Technical Specification (TS) 38.331, 3rd Generation Partnership Project (3GPP), 4 2019. Version 15.5.1.
[7] 3GPP. Nr;medium access control (mac) protocol specification (release 15). Technical Specification (TS) 38.321, 3rd Generation Partnership Project (3GPP), 4 2019. Version 15.5.0.
[8] 3GPP. Study on User Equipment (UE) power saving in NR. Technical report (TR) 38.840, 3rd Generation Partnership Project (3GPP), 06 2019. Version 16.0.0.
[9] A. Awada and D. S. Michalopoulos and A. Ali. An improved method for ondemand system information broadcast in 5G networks. In 2017 IEEE Conference on Standards for Communications and Networking (CSCN), pages 18–23, Sept 2017.
[10] M. Agiwal, A. Roy, and N. Saxena. Next generation 5g wireless networks: A comprehensive survey. IEEE Communications Surveys Tutorials, 18(3):1617–1655, thirdquarter 2016.
[11] M. Giordani, M. Mezzavilla, and M. Zorzi. Initial access in 5g mmwave cellular networks. IEEE Communications Magazine, 54(11):40–47, November 2016.
[12] M. A. Ingale and A. Agiwal. On demand system information delivery for 5G wireless system. In GLOBECOM 2017 2017 IEEE Global Communications Conference, pages 1–6, Dec 2017.
[13] C. Jeong, J. Park, and H. Yu. Random access in millimeterwave beamforming cellular networks: issues and approaches. IEEE Communications Magazine, 53(1):180–185, January 2015.
[14] F. Khan, Z. Pi, and S. Rajagopal. Millimeterwave mobile broadband with large scale spatial processing for 5g mobile communication. In 2012 50th Annual Allerton Conference on Communication, Control, and Computing (Allerton), pages 1517– 1523, Oct 2012.
[15] M. Koseoglu. Lower bounds on the LTEA average random access delay under massive M2M arrivals. IEEE Transactions on Communications, 64(5):2104–2115, May 2016.
[16] S. Lien, S. Shieh, Y. Huang, B. Su, Y. Hsu, and H. Wei. 5g new radio: Waveform, frame structure, multiple access, and initial access. IEEE Communications Magazine, 55(6):64–71, June 2017.
[17] G. Lin, S. Chang, and H. Wei. Estimation and adaptation for bursty LTE random access. IEEE Transactions on Vehicular Technology, 65(4):2560–2577, April 2016.
[18] H. ShokriGhadikolaei, C. Fischione, G. Fodor, P. Popovski, and M. Zorzi. Millimeter wave cellular networks: A mac layer perspective. IEEE Transactions on Communications, 63(10):3437–3458, Oct 2015.
[19] R. R. Tyagi, F. Aurzada, K. Lee, and M. Reisslein. Connection Establishment in LTEA Networks: Justification of Poisson Process Modeling. IEEE Systems Journal, 11(4):2383–2394, Dec 2017.