Loading...
|
Please use this identifier to cite or link to this item:
https://nccur.lib.nccu.edu.tw/handle/140.119/153372
|
| Title: | 基於圖神經網路的D2D通訊功率控制方法
Power Control for D2D Communication with Graph Neural Network |
| Authors: | 吳東霖
Wu, Tung-Lin |
| Contributors: | 張宏慶
Jang, Hung-Chin
吳東霖
Wu, Tung-Lin |
| Keywords: | 圖神經網路
圖嵌入
無線資源管理
功率控制
D2D通訊
Graph neural network
graph embedding
radio resource management
power control
D2D communication |
| Date: | 2024 |
| Issue Date: | 2024-09-04 14:58:30 (UTC+8) |
| Abstract: | 拜行動通訊快速發展之賜，連網裝置的數量不斷成長，然而裝置數量的增加也使得無線資源不敷使用，D2D通訊(Device-to-Device Communication)是一種用來減緩無線資源不足的技術，透過裝置間直接通訊以節省基地台用於轉發的無線資源，只是裝置間的通訊會互相干擾影響網路品質，因此需要基地台執行無線資源管理以提高資源使用效率。本研究改進無線資源管理中的功率控制算法，過往在模型導向的算法下，非凸優化的特性令功率控制的算法在效能與計算成本之間難以取得平衡，然而得益於資料導向的神經網路算法高速發展，兼具兩者的實時控制算法得以實現，至此眾多算法開始運用神經網路處理愈趨複雜的無線網路使用情景。最初的監督式學習依賴模型導向的算法結果，隨後透過更改為非監督式學習，使算法效能不再受限於標記資料，又由於裝置間的距離深刻影響無線訊號品質，算法轉向捕捉裝置的空間關係以提高效能，最後基於對模型適應複雜環境的需求，圖神經網路(Graph Neural Network, GNN)受到許多研究的重視。GNN擅長在低計算成本的限制下應對繁複的圖結構，故適合變換多端的無線環境，只是低計算成本也導致算法效能不如其它神經網路。在考量無線環境的訊號品質受鄰近裝置的影響後，本研究在GNN的基礎下，透過圖嵌入方法提高算法捕捉圖結構特徵的能力。為了驗證本研究的算法效能，通過實驗衡量算法適應不同環境的能力，同時也與其它GNN算法比較效能差異，實驗結果顯示，雖然計算時間相對較多，但本研究不僅在訓練環境與測試相同時有較好的效能，當訓練環境比測試環境複雜時，效能依舊能維持領先。
The rapid development of wireless communications increases the number of connected devices, resulting in a shortage of radio resources. Device-to-Device (D2D) communication alleviates the shortage through direct device communication. Nevertheless, it will cause interference and affect network quality. Therefore, radio resource management (RRM) is needed to enhance efficiency. This study aims to improve the power control algorithm in RRM. Previously, under the model-oriented algorithm, the non-convex optimization problem made it difficult to balance the performance and computational cost. However, the introduction of data-oriented neural network enabled real-time power control algorithms. Research started to use neural network to deal with RRM. Early algorithms used model-based results for supervised learning but were later shifted to unsupervised learning to overcome limitations of labeled data. Since the distance between devices profoundly affects signal quality, some algorithms try to capture spatial relationships to improve performance. Finally, due to the need to deal with complex environments, Graph Neural Network (GNN) have been employed as a solution. GNN excel at handling complex graph models with low computational costs but often lead to performance issues. By addressing the effects of nearby devices on signal quality, this study employs graph embedding methods to improve GNN’s ability to capture graph features. To verify the performance of the proposed algorithm, several experiments were conducted and compared with other GNN algorithms. Despite the relatively long computation time of the proposed algorithm, the experimental results indicate that it outperforms existing algorithms. |
| Reference: | [1] D. Feng, L. Lu, Y. Yuan-Wu, G. Y. Li, G. Feng and S. Li, "Device-to-Device Communications Underlaying Cellular Networks," in IEEE Transactions on Communications, vol. 61, no. 8, pp. 3541-3551, Aug. 2013.
[2] M. Noura and R. Nordin, "A survey on interference management for device-to-device (D2D) communication and its challenges in 5G Networks," Journal of Network and Computer Applications, vol. 71, pp. 130–150, 2016.
[3] Q. Shi, M. Razaviyayn, Z. -Q. Luo and C. He, "An Iteratively Weighted MMSE Approach to Distributed Sum-Utility Maximization for a MIMO Interfering Broadcast Channel," in IEEE Transactions on Signal Processing, vol. 59, no. 9, pp. 4331-4340, Sep. 2011.
[4] K. Shen and W. Yu, "FPLinQ: A cooperative spectrum sharing strategy for device-to-device communications," 2017 IEEE International Symposium on Information Theory (ISIT), Aachen, Germany, pp. 2323-2327, 2017.
[5] H. Sun, X. Chen, Q. Shi, M. Hong, X. Fu and N. D. Sidiropoulos, "Learning to Optimize: Training Deep Neural Networks for Interference Management," in IEEE Transactions on Signal Processing, vol. 66, no. 20, pp. 5438-5453, 15 Oct. 2018.
[6] D. Xu, X. Chen, C. Wu, S. Zhang, S. Xu and S. Cao, "Energy-Efficient Subchannel and Power Allocation for HetNets Based on Convolutional Neural Network," 2019 IEEE 89th Vehicular Technology Conference (VTC2019-Spring), Kuala Lumpur, Malaysia, pp. 1-5, 2019.
[7] F. Liang, C. Shen, W. Yu and F. Wu, "Towards Optimal Power Control via Ensembling Deep Neural Networks," in IEEE Transactions on Communications, vol. 68, no. 3, pp. 1760-1776, Mar. 2020.
[8] W. Lee, M. Kim and D. -H. Cho, "Deep Power Control: Transmit Power Control Scheme Based on Convolutional Neural Network," in IEEE Communications Letters, vol. 22, no. 6, pp. 1276-1279, June 2018.
[9] W. Cui, K. Shen and W. Yu, "Spatial Deep Learning for Wireless Scheduling," 2018 IEEE Global Communications Conference (GLOBECOM), Abu Dhabi, United Arab Emirates, pp. 1-6, 2018.
[10] M. Lee, G. Yu and G. Y. Li, "Graph Embedding-Based Wireless Link Scheduling With Few Training Samples," in IEEE Transactions on Wireless Communications, vol. 20, no. 4, pp. 2282-2294, Apr. 2021.
[11] Y. Shen, Y. Shi, J. Zhang and K. B. Letaief, "Graph Neural Networks for Scalable Radio Resource Management: Architecture Design and Theoretical Analysis," in IEEE Journal on Selected Areas in Communications, vol. 39, no. 1, pp. 101-115, Jan. 2021.
[12] Y. Shen, J. Zhang, S. H. Song and K. B. Letaief, "Graph Neural Networks for Wireless Communications: From Theory to Practice," in IEEE Transactions on Wireless Communications, vol. 22, no. 5, pp. 3554-3569, May 2023.
[13] T. S. Rappaport, "Wireless Communications: Principles and Practice," 2nd ed. Upper Saddle River, NJ, USA: Prentice Hall, 2002.
[14] Y. Gu, C. She, Z. Quan, C. Qiu and X. Xu, "Graph Neural Networks for Distributed Power Allocation in Wireless Networks: Aggregation Over-the-Air," in IEEE Transactions on Wireless Communications, vol. 22, no. 11, pp. 7551-7564, Nov. 2023.
[15] T. Chen, X. Zhang, M. You, G. Zheng and S. Lambotharan, "A GNN-Based Supervised Learning Framework for Resource Allocation in Wireless IoT Networks," in IEEE Internet of Things Journal, vol. 9, no. 3, pp. 1712-1724, 1 Feb. 2022.
[16] T. N. Kipf, M. Welling, "Semi-supervised classification with graph convolutional networks," in Proc. of ICLR, Apr. 2017.
[17] P. Velickovic, G. Cucurull, A. Casanova, A. Romero, P. Lio, and Y. Bengio, "Graph attention networks," in Proc. of ICLR, 2017.
[18] W. L. Hamilton, Z. Ying, and J. Leskovec, "Inductive representation learning on large graphs," in Proc. of NIPS, 2017.
[19] K. Xu, W. Hu, J. Leskovec, and S. Jegelka, "How powerful are graph neural networks," in Proc. of ICLR, 2019.
[20] W. Jiang, "Graph-based deep learning for communication networks: A survey," Comput. Commun., vol. 185, pp. 40-54, Mar. 2022.
[21] T. Mikolov, K. Chen, G. Corrado, and J. Dean, "Efficient Estimation of Word Representations in Vector Space," in Proc. of ICLR, Jan. 2013.
[22] B. Perozzi, R. Al-Rfou, and S. Skiena, "Deepwalk: Online learning of social representations," in Proc. of ACM, pp. 701–710, 2014.
[23] Horsmalahti, Panu, "Comparison of Bucket Sort and RADIX Sort," arXiv preprint, arXiv:1206.3511, Jun. 2012.
[24] J. Gilmer, S. S. Schoenholz, P. F. Riley, O. Vinyals, and G. E. Dahl, "Neural message passing for quantum chemistry," in Proc. of PMLR, pp. 1263-1272, Aug. 2017.
[25] Recommendation ITU-R P.1411-12. International Telecommunication Union, Aug. 2023. |
| Description: | 碩士
國立政治大學
資訊科學系
107753025 |
| Source URI: | http://thesis.lib.nccu.edu.tw/record/#G0107753025 |
| Data Type: | thesis |
| Appears in Collections: | [資訊科學系] 學位論文
|
Files in This Item:
| File |
Description |
Size | Format | |
|---|
| 302501.pdf | | 2933Kb | Adobe PDF | 0 | View/Open |
|
All items in 政大典藏 are protected by copyright, with all rights reserved.
|
