整合接取與後傳多段式高密度小蜂巢網路: 資源分配與小基站部署

Integrated-Access-and-Backhaul Multi-hop Dense Small Cell Networks: Resource Allocation and Node Placement

賴俊佑(Jiun-You Lai)

指導教授 ： 蘇育德

國立交通大學/電機學院/電信工程系所/博士(2020年)

摘要

以一個大型(主)基站(MBS)與若干個小型基站(SBS)為單位所組合而成的超高密度小蜂巢(ultra-dense small cell)異質網路是典型的5G毫米波移動寬頻通訊網路架構。大基站除了服務鄰近的用戶外，同時也借助多段傳送的方式連結小基站以服務距離較遠的用戶。整合接取與後傳(Integrated-Access-and-Backhaul, IAB)是一個新興的技術，在IAB網路，小基站透過無線鏈路同時擔任用戶接取端點與後傳的中繼站。本文旨在研究:當此無線網路配以多天線正交分頻多工(MIMO-OFDM)的傳輸技術時，如何分配無線資源及選擇小基站的位置。
首先我們發展一套最佳的資源分配演算法(基於加權總容量最大化方式)。其中資源分配包含了子載波(subcarrier)及空間子通道(subchannel)的指定以及相關的功率分配。我們另外提出兩種低複雜度的演算法並透過模擬驗證，所提的低複雜度演算法在效能表現上與最佳演算法比較，在高訊號雜音比時雖會有些微損失但計算複雜度卻可大大的降低。我們的資源分配演算法亦可應用於其他的用戶關聯(UE association)規則以及小基站佈署的多段式無線網路系統。另外，我們考慮了因為系統回報量測的通道狀態訊息給主基地台與此訊息實際用於傳輸的時間延遲而導致通道老化現象。我們展示了利用通道預測法來克服通道老化所帶來的效益以及集中式資源分配方法的限制。同時也探討了頻率再利用以及多段式架構的好處。

對於整合接取與後傳的蜂巢式網路，一個可能更重要的系統設計問題是小基站佈署問題。根據給定的用戶關聯規則和用戶位置之統計分佈，我們發展了一套系統性的方法來找到最佳小基站位置。針對兩種特殊的電波傳播模型，我們推導出小基站位置的閉合表示式，該小基站位置的佈署可以最大化頻譜效率的下限。電腦模擬數據與閉合式解決方案的一致驗證了我們估計的準確性。

關鍵字

整合接取與後傳 ； 資源分配 ； 小基站部署 ； 多天線正交分頻多工 ； 用戶關聯 ； 通道老化

並列摘要

In this dissertation, we consider a heterogeneous MIMO-OFDMA based dense small
cell (SC) system in which each macro cell base station (MBS) serves its coverage area
with the help of small cell base stations (SBSs) through multi-hop wireless connections.
The SBSs act as integrated access and backhaul (IAB) nodes that handle both access
and backhaul traffics with wireless links.
We first develop an optimal (sum-rate maximization) resource allocation (RA) algorithm which considers subcarriers/spatial subchannels assignment and the associated power allocation. We also present two low-complexity suboptimal RA schemes which, as verified by simulations, incur only minor performance loss in the high signal to noise ratio (SNR) region. Our RA algorithms can be applied to other multi-hop networks with general user equipment (UE) association rule and node location distributions. We study the channel aging effect caused by the time lag between the time channel state information (CSI) is measured and that when data transmission occurs. We show the benefit of channel prediction and the limit of a centralized RA approach. The advantages of frequency (channel) reuse and the multi-hop architecture are demonstrated as well.
A related but perhaps more important system design issue for an IAB cellular network is the IAB node placement problem. With the given UE association rule and UE location distribution, we present systematic approaches to find the optimal node locations. For two special propagation models, we derive closed-form expressions for the node locations that maximizes a spectral efficiency lower bound. Numerical results validate the accuracy of our estimates based on either numerical evaluations or closed-form solutions.

並列關鍵字

Integrated-Access-and-Backhaul ； Resource allocation ； Node placement ； MIMO-OFDM ； UE association ； Channel aging

參考文獻

Bibliography

[1] M. Ding, D. L.-P_erez, H. Claussen, and M. A. Kaafar, “On the fundamental characteristics of ultra-dense small cell networks," IEEE Network May/June 2018.

[2] S. A. Banani, et al., “Analyzing dependent placements of small cells in a twolayer heterogeneous network with a rate coverage constraint," IEEE Trans. Vehi. Technol., vol. 65, no. 12, pp. 9801-9816, Dec. 2016.

[3] X. Ge, et al., “5G ultra-dense cellular networks," IEEE Wireless Commun., vol. 23, no.2, pp.72-79, Feb. 2016.

[4] X. Ge, S. Tu, G. Mao, V. K. N. Lau, L. Pan, “Cost efficiency optimization of 5G wireless backhaul networks," IEEE Trans. Mobile Computing, to be published, 2019.