在信息傳輸過程中如何保證信息的安全性是通信中的重要問題之一。目前常用的保密傳輸方式是基於密鑰的加密技術，但是隨著現代無線通信網絡的發展和計算資源的不斷豐富，基於密鑰的技術在無線網絡中的應用正面臨着巨大挑戰。這些挑戰一方面來自於無線介質的開放性使得竊聽更為容易，另一方面由於動態無線網絡和自組織無線網絡的發展使得密鑰的發布和管理更為困難。因此，為解決密鑰技術所面臨的挑戰，基於物理層的保密傳輸技術研究在近些年受到了很大關注。該技術最早在七十年代由Wyner 提出，其核心思想是利用無線信道的隨機性和目標用戶與竊聽者間無線信道容量的差異，通過對發射信號進行編碼設計使得目標用戶能正確解碼信息但竊聽者卻不能。該技術的關鍵問題之一是如何對發射信號進行設計從而提高保密信息的傳輸速率（或保密容量）。本論文的主要工作即是以此出發，旨在研究不同通信場景下最優化發射信號的設計，具體而言，本論文主要研究了以下場景下的最優發射信號設計： / 本論文的第一部分考慮一個多天線的發射機傳輸保密信息給一個單天線的目標用戶，同時有多個多天線的偷聽者在偷聽的場景。我們的目標是設計最優化發射信號使得保密信息傳輸速率最大化。該優化問題的難點在於保密信息速率函數是發射信號的一個非凸函數，因而很難求解到全局最優解。我們通過運用凸鬆弛技術證明這個非凸優化問題的全局最優解可以由它的凸鬆弛問題得到，並且我們證明了最優化的發射信號方案是採用波束聚焦。以上結論在發射機完全知道和部分知道接收機的信道信息時均成立。 / 本論文的第二部分是在第一部分的基礎上考慮在發射信號中加入人為噪聲以輔助保密信息的傳輸。具體而言，發射機可以分配部分功率來發射人為噪聲以達到干擾竊聽者的接收的目的。儘管在現有很多研究中已經證明了這種人為噪聲輔助的發射方式可有效提高保密信息傳輸速率，但是如何對保密信號和人為噪聲進行聯合優化設計使得保密傳輸速率最大化的問題一直未能有效解決。在本論文中，我們給出了一種保密信號和人為噪聲聯合最優化的求解方案。該方案是基於優化理論中的半正定規劃算法來獲得全局最優解，並且該算法在發射機完全知道和部分知道接收機信道信息時均適用。 / 本論文的第三部分主要考慮的是發射機、目標用戶和偷聽者均是多天線情況下，最大化保密信息容量的發射信號設計問題。該優化問題可以看作是之前單天線目標用戶的一個推廣，但較之前的最優信號設計問題更加具有挑戰性。在目前已知的工作中，該優化問題還沒有一個有效的多項式時間算法能求解到全局最優解。這裡，我們提出了一種基於交替優化算法的發射信號設計方案來獲得（局部）最優發射信號設計。我們證明該交替優化算法可以通過迭代注水算法來實現，因而具有很低的複雜度，並且該算法可以保證收斂到原最優化問題的穩定點，因而可以在多數情況下獲得（局部）最優解。同時在該部分，我們也研究了在發射機部分知道信道信息狀態時魯棒性發射信號的設計問題，並給出了基於交替優化算法的魯棒發射信號設計。 / 除以上提到的主要結果，本論文還考慮了多播保密信息速率最優化發射信號設計，和具有中斷概率約束的保密信息速率最優化發射信號的設計。 / Security is one of the most important issues in communications. Conventional techniques for achieving confidentiality in communication networks are based on cryptographic encryption. However, for wireless networks, this technique is faced with more challenges due to the open nature of the wireless medium as well as the dynamic topology of mobile networks. In the 1970's, Wyner proposed a physical layer-based approach to achieve perfectly secure communication without using encryption. One of the key problems of Wyner's approach is how to optimally design the transmit signal such that a high secrecy rate (i.e., the data rate at which the confidential information can be securely transmitted) can be achieved. In our work, we aim to solve this transmit signal optimization problem under various scenarios using convex optimization techniques. Specifically, the thesis consists of the following three main parts: / In the first part, we consider a multi-input single-output (MISO) scenario, where a multi-antenna transmitter sends confidential information to a singleantenna legitimate receiver, in the presence of multiple multi-antenna eavesdroppers. Our goal is to maximize an achievable secrecy rate by appropriately designing the transmit signal. The challenge of this secrecy rate maximization (SRM) problem is that it is a nonconvex optimization problem by nature. We show, by convex relaxation, that this seemingly nonconvex SRM problem admits a convex equivalent under both perfect and imperfect channel state information (CSI) cases. Our result also indicates that transmit beamforming is an optimal transmit strategy, irrespective of the number of eavesdroppers and the number of antennas employed by each eavesdropper. This provides a useful design guideline for practical implementations. / In the second part, we consider a scenario where the transmitter is able to simultaneously generate artificial noise (AN) to interfere the eavesdroppers during the transmission of the confidential message. While the efficacy of AN in improving the system security has been demonstrated in many existing works, how to jointly optimize the AN and the transmit signal is still an unsolved problem. In this part, we solve this AN-aided SRM problem under the same scenario as the first part, and give an efficient, semidefinite program (SDP)- based line search approach to obtain an optimal transmit signal and AN design under both perfect and imperfect-CSI situations. / In the last part, we consider a secrecy capacity maximization (SCM) problem for a multi-input multi-output (MIMO) scenario, where the legitimate receiver and the eavesdropper are both equipped with multiple antennas. This MIMOSCM problem is a generalization of the previous MISO-SRM problems. So far there is no known efficient algorithm to solve this problem in a global optimal manner. Herein, we propose an alternating optimization algorithm to tackle the SCM problem. The proposed algorithm has a nice iterative water-filling interpretation and is guaranteed to converge to a stationary solution of the MIMO-SCM problem. Extensions to robust SCM are also investigated in this part. / Besides the above three main results, this thesis also developed some approximate solutions to the multicast SRM of multiple MISO legitimate channels overheard by multiple MIMO eavesdroppers, and to the outage-constrained SRM of an MISO legitimate channel overheard by multiple MISO eavesdroppers. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Detailed summary in vernacular field only. / Li, Qiang. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2012. / Includes bibliographical references. / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Contributions of This Thesis --- p.3 / Chapter 1.2 --- Organization of This Thesis --- p.5 / Chapter 2 --- Fundamentals of Physical-Layer Secrecy --- p.6 / Chapter 2.1 --- Elements of Information Theoretic Security --- p.6 / Chapter 2.2 --- Transmit Design for Physical-layer Secrecy: State-of-the-Art --- p.14 / Chapter 2.2.1 --- MISO Secrecy Capacity Maximization --- p.14 / Chapter 2.2.2 --- MIMO Secrecy Capacity Maximization --- p.17 / Chapter 2.2.3 --- AN-aided Secrecy Rate Maximization --- p.21 / Chapter 2.2.4 --- Secrecy Rate Maximization with Additional Covariance Constraints --- p.24 / Chapter 2.2.5 --- Robust Transmit Design for Physical-Layer Secrecy under Imperfect CSI --- p.28 / Chapter 2.3 --- Summary --- p.36 / Chapter 2.4 --- Appendix: GSVD --- p.37 / Chapter 3 --- MISOMEs Secrecy Rate Maximization --- p.38 / Chapter 3.1 --- System Model and Problem Statement --- p.39 / Chapter 3.1.1 --- System Model --- p.39 / Chapter 3.1.2 --- Problem Statement --- p.40 / Chapter 3.2 --- An SDP Approach to SRM Problem (3.4) --- p.43 / Chapter 3.2.1 --- The Secrecy-Rate Constrained Problem --- p.44 / Chapter 3.2.2 --- The Secrecy-Rate Maximization Problem --- p.46 / Chapter 3.3 --- Secrecy-Rate Optimization with Imperfect CSI --- p.49 / Chapter 3.3.1 --- Robust Secrecy-Rate Problem Formulations --- p.49 / Chapter 3.3.2 --- The Robust Secrecy-Rate Constrained Problem --- p.50 / Chapter 3.3.3 --- The Robust Secrecy-Rate Maximization Problem --- p.53 / Chapter 3.4 --- Simulation Results --- p.55 / Chapter 3.4.1 --- The Perfect CSI Case --- p.56 / Chapter 3.4.2 --- The Imperfect CSI Case --- p.58 / Chapter 3.5 --- Summary --- p.59 / Chapter 3.6 --- Appendix --- p.61 / Chapter 3.6.1 --- Proof of Proposition 3.1 --- p.61 / Chapter 3.6.2 --- Verifying Slater's Constraint Qualification for Problem (3.10) --- p.63 / Chapter 3.6.3 --- Proof of Theorem 3.1 --- p.63 / Chapter 3.6.4 --- Relationship between Super-Eve Design and the Optimal SDP Design --- p.65 / Chapter 3.6.5 --- Proof of Proposition 3.4 --- p.67 / Chapter 3.6.6 --- Proof of Proposition 3.5 --- p.70 / Chapter 3.6.7 --- Worst-case Secrecy Rate Calculation --- p.71 / Chapter 4 --- Multicast Secrecy Rate Maximization --- p.73 / Chapter 4.1 --- System Model and Problem Statement --- p.74 / Chapter 4.2 --- An SDP Approximation to Problem (4.2) --- p.75 / Chapter 4.3 --- Simulation Results --- p.78 / Chapter 4.4 --- Summary --- p.79 / Chapter 4.5 --- Appendix --- p.81 / Chapter 4.5.1 --- Proof of Proposition 4.1 --- p.81 / Chapter 4.5.2 --- Proof of Theorem 4.1 --- p.82 / Chapter 5 --- AN-aided MISOMEs Secrecy Rate Maximization --- p.85 / Chapter 5.1 --- System Model and Problem Statement --- p.86 / Chapter 5.1.1 --- System Model --- p.86 / Chapter 5.1.2 --- Problem Statement --- p.87 / Chapter 5.2 --- An SDP-based Approach to the SRM Problem --- p.89 / Chapter 5.2.1 --- A Tight Relaxation of the SRM Problem (5.4) --- p.90 / Chapter 5.2.2 --- An SDP-based Line Search Method for the Relaxed SRM Problem (5.9) --- p.92 / Chapter 5.3 --- Robust Transmit Design for Worst-Case SRM --- p.94 / Chapter 5.3.1 --- Worst-Case Robust SRM Problem Formulation --- p.95 / Chapter 5.3.2 --- A Tight Relaxation of the WCR-SRM Problem (5.17) --- p.96 / Chapter 5.3.3 --- An SDP-based Line Search Method for the RelaxedWCRSRM Problem (5.23) --- p.98 / Chapter 5.4 --- Robust Transmit Design for Outage SRM --- p.100 / Chapter 5.4.1 --- A Sphere-bounding Safe Approximation to OCR-SRM Problem (5.29) --- p.101 / Chapter 5.5 --- Simulation Results --- p.103 / Chapter 5.5.1 --- The Perfect-CSI Case --- p.104 / Chapter 5.5.2 --- The Imperfect-CSI: Bounded Spherical Uncertainty --- p.105 / Chapter 5.5.3 --- The Imperfect-CSI: Gaussian Random Uncertainty --- p.108 / Chapter 5.6 --- Summary --- p.111 / Chapter 5.7 --- Appendix --- p.112 / Chapter 5.7.1 --- Proof of Proposition 5.1 --- p.112 / Chapter 5.7.2 --- Proof of Theorem 5.1 --- p.114 / Chapter 5.7.3 --- Proof of Theorem 5.2 --- p.117 / Chapter 6 --- Outage Secrecy Rate Maximization for MISOSEs --- p.120 / Chapter 6.1 --- System Model and Problem Statement --- p.121 / Chapter 6.2 --- A Bernstein-type Inequality-Based Safe Approximation to Problem (6.2) --- p.122 / Chapter 6.3 --- Simulation Results --- p.127 / Chapter 6.4 --- Summary --- p.128 / Chapter 6.5 --- Appendix --- p.129 / Chapter 6.5.1 --- Proof of Lemma 6.1 --- p.129 / Chapter 6.5.2 --- Proof of Proposition 6.1 --- p.130 / Chapter 7 --- MIMOME Secrecy Rate Maximization --- p.134 / Chapter 7.1 --- An Alternating Optimization Approach to the MIMO-SCM Problem (7.1) --- p.135 / Chapter 7.2 --- An Alternating Optimization Approach to theWorst-case MIMOSCM Problem --- p.140 / Chapter 7.3 --- An Alternating Optimization Approach to the Outageconstrained SCM --- p.142 / Chapter 7.4 --- Simulation Results --- p.145 / Chapter 7.4.1 --- The Perfect CSI Case --- p.146 / Chapter 7.4.2 --- The Imperfect CSI case --- p.149 / Chapter 7.5 --- Summary --- p.150 / Chapter 7.6 --- Appendix --- p.153 / Chapter 7.6.1 --- Proof of Proposition 7.1 --- p.153 / Chapter 7.6.2 --- Proof of the monotonicity of Tr(W? ) w.r.t. --- p.154 / Chapter 7.6.3 --- Proof of Proposition 7.2 --- p.155 / Chapter 8 --- Conclusion --- p.157 / Chapter 8.1 --- Summary --- p.157 / Chapter 8.2 --- Future Directions --- p.158 / Bibliography --- p.161
| Identifer | oai:union.ndltd.org:cuhk.edu.hk/oai:cuhk-dr:cuhk_328097 |
| Date | January 2012 |
| Contributors | Li, Qiang, Chinese University of Hong Kong Graduate School. Division of Electronic Engineering. |
| Source Sets | The Chinese University of Hong Kong |
| Language | English, Chinese |
| Detected Language | English |
| Type | Text, bibliography |
| Format | electronic resource, electronic resource, remote, 1 online resource (1 v. (various pagings)) : ill. (some col.) |
| Rights | Use of this resource is governed by the terms and conditions of the Creative Commons “Attribution-NonCommercial-NoDerivatives 4.0 International” License (http://creativecommons.org/licenses/by-nc-nd/4.0/) |
Page generated in 0.0081 seconds