指導教授：謝宏昀臺灣大學：電信工程學研究所許晉瑋Hsu, Chin-WeiChin-WeiHsu2014-11-302018-07-052014-11-302018-07-052014http://ntur.lib.ntu.edu.tw//handle/246246/264316基於IEEE 802.11的Wi-Fi無線通訊技術發展迅速，目前幾乎每一個家庭都裝有一個以上的Wi-Fi基地台。 隨處可見的Wi-Fi基地台雖然提供相當高的覆蓋率，但因IEEE 802.11分散式協同功能(DCF)屬競爭式演算法，在高密度佈建下，將會面對來自鄰近基地台信號干擾、隱藏節點等問題。 為了解決這些問題，Wi-Fi基地台之間的協同式合作將是提高無線網路效率的重要方法，而功率控制更是其中解決信號干擾等問題的關鍵技術。 雖然相關文獻上已有許多探討功率控制的模 型，但他們並未準確地考慮到分散式協同功能。 在本論文中，我們首先分析與建構DCF成功傳輸模型，包含成功通道競爭、資料傳收與ACK傳收等三個步驟。 透過定義並建構不同的集合，我們可以準確地描述Wi-Fi基地台在傳收過程中對彼此的影響。 基於本論文所建構出的DCF成功傳輸模型，我們進一步提出功率控制最佳化設計，包含其最佳化模型與演算法，以提高一個Wi-Fi網路下最差使用者的成功傳輸機率。 模擬結果顯示，相較於傳統的功率控制模型，本論文提出之模型有更高的成功傳輸機率以及更佳的網路公平性。 從微觀的模擬結果我們亦可發現干擾、隱藏節點等問題嚴重影響了傳統功率控制模型的網路品質，而我們的模型能精確地分析與考慮分散式協調功能的特性，可以有效地解決Wi-Fi高密度佈建下的問題。Due to the rapid development of Wi-Fi networks based on the IEEE 802.11 standards, almost every household has installed at least one Wi-Fi access point(AP). Although dense deployment of APs provides high coverage of wireless networks, high power level of signal from neighbor APs causes many problems including interference and hidden terminals resulted from the contention-based distributed coordinated function (DCF) used by 802.11. Therefore, the collaborative mechanisms between APs will be one of important methods to improve the performance of wireless network and power control will especially be key technique to solve these problems. Although there are many researches discussing about power control model, they do not consider the behavior of DCF. In this thesis, we first analyze the characteristics of DCF and model successful transmission for DCF including three steps: accessing channel, data reception, and ACK reception. We define different sets which can accurately represent the interaction and influence between APs for three steps during transmission. Based on the model of successful transmission for DCF, we find that power setting indeed influences the performance of 802.11 network and we propose power control design. The simulation results show that proposed power control design outperform other models for not only probability of successful transmission but also fairness of network. From microscopic view of simulation results, we find that traditional models suffer from problems of hidden terminal and high interference. For our proposed design, these problems can be solved effectively because we consider and analyze the characteristics of DCF precisely.ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii LIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v LIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi CHAPTER 1 INTRODUCTION . . . . . . . . . . . . . . . . . . . . 1 CHAPTER 2 BACKGROUND AND RELATED WORK . . . . . 4 2.1 802.11 Distributed Coordination Function . . . . . . . . . . . . . . 4 2.2 Analysis of DCF . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 2.3 Cooperated System of Wi-Fi APs . . . . . . . . . . . . . . . . . . 11 2.4 Power Control and Interference Management . . . . . . . . . . . . 13 CHAPTER 3 PROPOSED MODEL FOR CHARACTERISTICS OF DISTRIBUTED COORDINATED FUNCTION . . . . . . . 16 3.1 Definition of Different Sets . . . . . . . . . . . . . . . . . . . . . . 17 3.1.1 Carrier Sense Set and Silence Set . . . . . . . . . . . . . . . 17 3.1.2 Interference Set . . . . . . . . . . . . . . . . . . . . . . . . 19 3.2 Three Essential Steps for Successful Transmission . . . . . . . . . 21 3.2.1 Step 1: Successful Accessing Channel . . . . . . . . . . . . 22 3.2.2 Step 2: Successful Data Reception . . . . . . . . . . . . . . 23 3.2.3 Step 3: Successful ACK Reception . . . . . . . . . . . . . . 26 3.3 Validation of Three Essential Steps for Successful Transmission . . 28 3.3.1 Validation of Successful Access Channel . . . . . . . . . . . 28 3.3.2 Validation of Data Reception . . . . . . . . . . . . . . . . . 31 3.3.3 Validation of ACK reception . . . . . . . . . . . . . . . . . 32 CHAPTER 4 NETWORK SCENARIO AND FORMULATION OF POWER CONTROL MODEL . . . . . . . . . . . . . . . . . . 34 4.1 Network Scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 4.2 Formulation of Constraints . . . . . . . . . . . . . . . . . . . . . . 35 4.3 Formulation of Objective Function . . . . . . . . . . . . . . . . . . 37 4.3.1 Original Objective Function . . . . . . . . . . . . . . . . . 37 4.3.2 Approximation of Objective Function . . . . . . . . . . . . 39 4.3.3 Formulation of Power Control Problem . . . . . . . . . . . 40 4.4 Relaxation of Original Optimization Problem . . . . . . . . . . . . 42 4.4.1 Relaxing the Zero-One Function . . . . . . . . . . . . . . . 42 4.4.2 Relaxing Minimization of the Maximum Value . . . . . . . 43 CHAPTER 5 PROPOSED ALGORITHM AND EVALUATION 46 5.1 Branch and Bound Algorithm . . . . . . . . . . . . . . . . . . . . 46 5.1.1 Proposed BB Algorithm . . . . . . . . . . . . . . . . . . . . 48 5.1.2 Complexity Discussion . . . . . . . . . . . . . . . . . . . . 50 5.2 Different Power Models . . . . . . . . . . . . . . . . . . . . . . . . 54 5.2.1 Traditional Exact Power Model . . . . . . . . . . . . . . . . 54 5.2.2 Iterative Power Update Model . . . . . . . . . . . . . . . . 55 5.3 Performance of the Proposed Model . . . . . . . . . . . . . . . . . 56 5.3.1 Simulation Results of Psuc, VCS . . . . . . . . . . . . . . . . 57 5.3.2 Microscopic View of Simulation Results . . . . . . . . . . . 61 5.3.3 Fairness Discussion of Simulation Results . . . . . . . . . . 63 CHAPTER 6 INSIGHTS FROM SIMULATION RESULTS . . . 68 6.1 Intelligent Power Control Design . . . . . . . . . . . . . . . . . . . 68 6.1.1 Wi-Fi IPU Model . . . . . . . . . . . . . . . . . . . . . . . 68 6.1.2 Simulation Results of Wi-Fi IPU Model . . . . . . . . . . . 70 6.2 Adaption to Network Change . . . . . . . . . . . . . . . . . . . . . 72 6.2.1 Power Update Algorithm Adapting to Network Change . . 73 6.2.2 Simulation Results of Network Change . . . . . . . . . . . 74 CHAPTER 7 CONCLUSION AND FUTURE WORK . . . . . . 78 7.1 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 7.2 Possible Extensions . . . . . . . . . . . . . . . . . . . . . . . . . . 78 APPENDIX A — MODELING DCF FOR DIFFERENT SCENARIOS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 APPENDIX B — PROXIMAL INFORMATION EXCHANGE . 83 REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 871080070 bytesapplication/pdf論文公開時間：2016/08/21論文使用權限：同意有償授權(權利金給回饋學校)802.11基地台高密度佈建功率控制最佳化設計thesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/264316/1/ntu-103-R01942030-1.pdf