Robert C. Qiu, Department of Electrical and Computer Engineering, Tennessee Technological University, USAProfessor Qiu is currently Director of the Wireless Networking System Laboratory at Tennessee Technological University, USA. He was Founder-CEO and President of Wiscom Technologies, Inc., manufacturing and marketing WCDMA chipsets. Wiscom was acquired by Intel in 2003. Prior to Wiscom, he worked for GTE Labs, Inc. (now Verizon), Waltham, MA, and Bell Labs, Lucent, Whippany, NJ. He holds 5 U.S. patents (another two pending) in WCDMA. He is co-editor of Ultra-Wideband Wireless Communications and Networks (Wiley), and? has authored over 80 technical (journal/conference) papers and contributed 6 book chapters. Professor Qiu has contributed to 3GPP and IEEE standards bodies, and delivered invited seminars to institutions including Princeton University and the U.S. Army Research Lab. Dr. Qiu serves as Associate Editor, IEEE Transaction on Vehicular Technology, International Journal of Sensor Networks and Wireless Communication and Mobile Computing.

Select Guide Rating

• Preface xv1 Introduction 11.1 Vision: “Big Data” 11.2 Cognitive Radio: System Concepts 21.3 Spectrum Sensing Interface and Data Structures 21.4 Mathematical Machinery 41.4.1 Convex Optimization 41.4.2 Game Theory 61.4.3 “Big Data” Modeled as Large Random Matrices 61.5 Sample Covariance Matrix 101.6 Large Sample Covariance Matrices of Spiked Population Models 111.7 Random Matrices and Noncommutative Random Variables 121.8 Principal Component Analysis 131.9 Generalized Likelihood Ratio Test (GLRT) 131.10 Bregman Divergence for Matrix Nearness 132 Spectrum Sensing: Basic Techniques 152.1 Challenges 152.2 Energy Detection: No Prior Information about Deterministic or Stochastic Signal 152.2.1 Detection in White Noise: Lowpass Case 162.2.2 Time-Domain Representation of the Decision Statistic 192.2.3 Spectral Representation of the Decision Statistic 192.2.4 Detection and False Alarm Probabilities over AWGN Channels 202.2.5 Expansion of Random Process in Orthonormal Series with Uncorrelated Coefficients: The Karhunen-Loeve Expansion 212.3 Spectrum Sensing Exploiting Second-Order Statistics 232.3.1 Signal Detection Formulation 232.3.2 Wide-Sense Stationary Stochastic Process: Continuous-Time 242.3.3 Nonstationary Stochastic Process: Continuous-Time 252.3.4 Spectrum Correlation-Based Spectrum Sensing for WSS Stochastic Signal: Heuristic Approach 292.3.5 Likelihood Ratio Test of Discrete-Time WSS Stochastic Signal 322.3.6 Asymptotic Equivalence between Spectrum Correlation and Likelihood Ratio Test 352.3.7 Likelihood Ratio Test of Continuous-Time Stochastic Signals in Noise: Selin’s Approach 362.4 Statistical Pattern Recognition: Exploiting Prior Information about Signal through Machine Learning 392.4.1 Karhunen-Loeve Decomposition for Continuous-Time Stochastic Signal 392.5 Feature Template Matching 422.6 Cyclostationary Detection 473 Classical Detection 513.1 Formalism of Quantum Information 513.2 Hypothesis Detection for Collaborative Sensing 513.3 Sample Covariance Matrix 553.3.1 The Data Matrix 563.4 Random Matrices with Independent Rows 633.5 The Multivariate Normal Distribution 673.6 Sample Covariance Matrix Estimation and Matrix Compressed Sensing 773.6.1 The Maximum Likelihood Estimation 813.6.2 Likelihood Ratio Test (Wilks Test) for Multisample Hypotheses 833.7 Likelihood Ratio Test 843.7.1 General Gaussian Detection and Estimator-Correlator Structure 843.7.2 Tests with Repeated Observations 903.7.3 Detection Using Sample Covariance Matrices 943.7.4 GLRT for Multiple Random Vectors 953.7.5 Linear Discrimination Functions 973.7.6 Detection of Correlated Structure for Complex Random Vectors 984 Hypothesis Detection of Noncommutative Random Matrices 1014.1 Why Noncommutative Random Matrices? 1014.2 Partial Orders of Covariance Matrices: A < B 1024.3 Partial Ordering of Completely Positive Mappings: (A) < (B) 1044.4 Partial Ordering of Matrices Using Majorization: A ≺ B 1054.5 Partial Ordering of Unitarily Invariant Norms: |||A||| < |||B||| 1094.6 Partial Ordering of Positive Definite Matrices of Many Copies: K k=1 Ak ≤ K k=1 Bk 1094.7 Partial Ordering of Positive Operator Valued Random Variables: Prob(A ≤ X ≤ B) 1104.8 Partial Ordering Using Stochastic Order: A ≤st B 1154.9 Quantum Hypothesis Detection 1154.10 Quantum Hypothesis Testing for Many Copies 1185 Large Random Matrices 1195.1 Large Dimensional Random Matrices: Moment Approach, Stieltjes Transform and Free Probability 1195.2 Spectrum Sensing Using Large Random Matrices 1215.2.1 System Model 1215.2.2 Marchenko-Pastur Law 1245.3 Moment Approach 1295.3.1 Limiting Spectral Distribution 1305.3.2 Limits of Extreme Eigenvalues 1335.3.3 Convergence Rates of Spectral Distributions 1365.3.4 Standard Vector-In, Vector-Out Model 1375.3.5 Generalized Densities 1385.4 Stieltjes Transform 1395.4.1 Basic Theorems 1435.4.2 Large Random Hankel, Markov and Toepltiz Matrices 1495.4.3 Information Plus Noise Model of Random Matrices 1525.4.4 Generalized Likelihood Ratio Test Using Large Random Matrices 1575.4.5 Detection of High-Dimensional Signals in White Noise 1645.4.6 Eigenvalues of (A + B)−1B and Applications 1695.4.7 Canonical Correlation Analysis 1715.4.8 Angles and Distances between Subspaces 1735.4.9 Multivariate Linear Model 1735.4.10 Equality of Covariance Matrices 1745.4.11 Multiple Discriminant Analysis 1745.5 Case Studies and Applications 1755.5.1 Fundamental Example of Using Large Random Matrix 1755.5.2 Stieltjes Transform 1775.5.3 Free Deconvolution 1785.5.4 Optimal Precoding of MIMO Systems 1785.5.5 Marchenko and Pastur’s Probability Distribution 1795.5.6 Convergence and Fluctuations Extreme Eigenvalues 1805.5.7 Information plus Noise Model and Spiked Models 1805.5.8 Hypothesis Testing and Spectrum Sensing 1835.5.9 Energy Estimation in a Wireless Network 1855.5.10 Multisource Power Inference 1875.5.11 Target Detection, Localization, and Reconstruction 1875.5.12 State Estimation and Malignant Attacker in the Smart Grid 1915.5.13 Covariance Matrix Estimation 1935.5.14 Deterministic Equivalents 1975.5.15 Local Failure Detection and Diagnosis 2005.6 Regularized Estimation of Large Covariance Matrices 2005.6.1 Regularized Covariance Estimates 2015.6.2 Banding the Inverse 2035.6.3 Covariance Regularization by Thresholding 2045.6.4 Regularized Sample Covariance Matrices 2065.6.5 Optimal Rates of Convergence for Covariance Matrix Estimation 2085.6.6 Banding Sample Autocovariance Matrices of Stationary Processes 2115.7 Free Probability 2135.7.1 Large Random Matrices and Free Convolution 2185.7.2 Vandermonde Matrices 2215.7.3 Convolution and Deconvolution with Vandermonde Matrices 2295.7.4 Finite Dimensional Statistical Inference 2326 Convex Optimization 2356.1 Linear Programming 2376.2 Quadratic Programming 2386.3 Semidefinite Programming 2396.4 Geometric Programming 2396.5 Lagrange Duality 2416.6 Optimization Algorithm 2426.6.1 Interior Point Methods 2426.6.2 Stochastic Methods 2436.7 Robust Optimization 2446.8 Multiobjective Optimization 2486.9 Optimization for Radio Resource Management 2496.10 Examples and Applications 2506.10.1 Spectral Efficiency for Multiple Input Multiple Output Ultra-Wideband Communication System 2506.10.2 Wideband Waveform Design for Single Input Single Output Communication System with Noncoherent Receiver 2566.10.3 Wideband Waveform Design for Multiple Input Single Output Cognitive Radio 2626.10.4 Wideband Beamforming Design 2686.10.5 Layering as Optimization Decomposition for Cognitive Radio Network 2726.11 Summary 2827 Machine Learning 2837.1 Unsupervised Learning 2887.1.1 Centroid-Based Clustering 2887.1.2 k-Nearest Neighbors 2897.1.3 Principal Component Analysis 2897.1.4 Independent Component Analysis 2907.1.5 Nonnegative Matrix Factorization 2917.1.6 Self-Organizing Map 2927.2 Supervised Learning 2937.2.1 Linear Regression 2937.2.2 Logistic Regression 2947.2.3 Artificial Neural Network 2947.2.4 Decision Tree Learning 2947.2.5 Naive Bayes Classifier 2957.2.6 Support Vector Machines 2957.3 Semisupervised Learning 2987.3.1 Constrained Clustering 2987.3.2 Co-Training 2987.3.3 Graph-Based Methods 2997.4 Transductive Inference 2997.5 Transfer Learning 2997.6 Active Learning 2997.7 Reinforcement Learning 3007.7.1 Q-Learning 3007.7.2 Markov Decision Process 3017.7.3 Partially Observable MDPs 3027.8 Kernel-Based Learning 3037.9 Dimensionality Reduction 3047.9.1 Kernel Principal Component Analysis 3057.9.2 Multidimensional Scaling 3077.9.3 Isomap 3087.9.4 Locally-Linear Embedding 3087.9.5 Laplacian Eigenmaps 3097.9.6 Semidefinite Embedding 3097.10 Ensemble Learning 3117.11 Markov Chain Monte Carlo 3127.12 Filtering Technique 3137.12.1 Kalman Filtering 3147.12.2 Particle Filtering 3187.12.3 Collaborative Filtering 3197.13 Bayesian Network 3207.14 Summary 3218 Agile Transmission Techniques (I): Multiple Input Multiple Output 3238.1 Benefits of MIMO 3238.1.1 Array Gain 3238.1.2 Diversity Gain 3238.1.3 Multiplexing Gain 3248.2 Space Time Coding 3248.2.1 Space Time Block Coding 3258.2.2 Space Time Trellis Coding 3268.2.3 Layered Space Time Coding 3268.3 Multi-User MIMO 3278.3.1 Space-Division Multiple Access 3278.3.2 MIMO Broadcast Channel 3288.3.3 MIMO Multiple Access Channel 3308.3.4 MIMO Interference Channel 3318.4 MIMO Network 3348.5 MIMO Cognitive Radio Network 3368.6 Summary 3379 Agile Transmission Techniques (II): Orthogonal Frequency Division Multiplexing 3399.1 OFDM Implementation 3399.2 Synchronization 3419.3 Channel Estimation 3439.4 Peak Power Problem 3459.5 Adaptive Transmission 3459.6 Spectrum Shaping 3479.7 Orthogonal Frequency Division Multiple Access 3479.8 MIMO OFDM 3499.9 OFDM Cognitive Radio Network 3499.10 Summary 35010 Game Theory 35110.1 Basic Concepts of Games 35110.1.1 Elements of Games 35110.1.2 Nash Equilibrium: Definition and Existence 35210.1.3 Nash Equilibrium: Computation 35410.1.4 Nash Equilibrium: Zero-Sum Games 35510.1.5 Nash Equilibrium: Bayesian Case 35510.1.6 Nash Equilibrium: Stochastic Games 35610.2 Primary User Emulation Attack Games 36010.2.1 PUE Attack 36010.2.2 Two-Player Case: A Strategic-Form Game 36110.2.3 Game in Queuing Dynamics: A Stochastic Game 36210.3 Games in Channel Synchronization 36810.3.1 Background of the Game 36810.3.2 System Model 36810.3.3 Game Formulation 36910.3.4 Bayesian Equilibrium 37010.3.5 Numerical Results 37110.4 Games in Collaborative Spectrum Sensing 37210.4.1 False Report Attack 37310.4.2 Game Formulation 37310.4.3 Elements of Game 37410.4.4 Bayesian Equilibrium 37610.4.5 Numerical Results 37911 Cognitive Radio Network 38111.1 Basic Concepts of Networks 38111.1.1 Network Architecture 38111.1.2 Network Layers 38211.1.3 Cross-Layer Design 38411.1.4 Main Challenges in Cognitive Radio Networks 38411.1.5 Complex Networks 38511.2 Channel Allocation in MAC Layer 38611.2.1 Problem Formulation 38611.2.2 Scheduling Algorithm 38711.2.3 Solution 38911.2.4 Discussion 39011.3 Scheduling in MAC Layer 39111.3.1 Network Model 39111.3.2 Goal of Scheduling 39311.3.3 Scheduling Algorithm 39311.3.4 Performance of the CNC Algorithm 39511.3.5 Distributed Scheduling Algorithm 39611.4 Routing in Network Layer 39611.4.1 Challenges of Routing in Cognitive Radio 39711.4.2 Stationary Routing 39811.4.3 Dynamic Routing 40211.5 Congestion Control in Transport Layer 40411.5.1 Congestion Control in Internet 40411.5.2 Challenges in Cognitive Radio 40511.5.3 TP-CRAHN 40611.5.4 Early Start Scheme 40811.6 Complex Networks in Cognitive Radio 41711.6.1 Brief Introduction to Complex Networks 41811.6.2 Connectivity of Cognitive Radio Networks 42111.6.3 Behavior Propagation in Cognitive Radio Networks 42312 Cognitive Radio Network as Sensors 42712.1 Intrusion Detection by Machine Learning 42912.2 Joint Spectrum Sensing and Localization 42912.3 Distributed Aspect Synthetic Aperture Radar 42912.4 Wireless Tomography 43312.5 Mobile Crowdsensing 43412.6 Integration of 3S 43512.7 The Cyber-Physical System 43512.8 Computing 43612.8.1 Graphics Processor Unit 43712.8.2 Task Distribution and Load Balancing 43712.9 Security and Privacy 43812.10 Summary 438Appendix A Matrix Analysis 441A.1 Vector Spaces and Hilbert Space 441A.2 Transformations 443A.3 Trace 444A.4 Basics of C ∗-Algebra 444A.5 Noncommunicative Matrix-Valued Random Variables 445A.6 Distances and Projections 447A.6.1 Matrix Inequalities 450A.6.2 Partial Ordering of Positive Semidefinite Matrices 451A.6.3 Partial Ordering of Hermitian Matrices 451References 453Index 511

Robert C. Qiu, Department of Electrical and Computer Engineering, Tennessee Technological University, USAProfessor Qiu is currently Director of the Wireless Networking System Laboratory at Tennessee Technological University, USA. He was Founder-CEO and President of Wiscom Technologies, Inc., manufacturing and marketing WCDMA chipsets. Wiscom was acquired by Intel in 2003. Prior to Wiscom, he worked for GTE Labs, Inc. (now Verizon), Waltham, MA, and Bell Labs, Lucent, Whippany, NJ. He holds 5 U.S. patents (another two pending) in WCDMA. He is co-editor of Ultra-Wideband Wireless Communications and Networks (Wiley), and? has authored over 80 technical (journal/conference) papers and contributed 6 book chapters. Professor Qiu has contributed to 3GPP and IEEE standards bodies, and delivered invited seminars to institutions including Princeton University and the U.S. Army Research Lab. Dr. Qiu serves as Associate Editor, IEEE Transaction on Vehicular Technology, International Journal of Sensor Networks and Wireless Communication and Mobile Computing.

Select Guide Rating

• Preface xv1 Introduction 11.1 Vision: “Big Data” 11.2 Cognitive Radio: System Concepts 21.3 Spectrum Sensing Interface and Data Structures 21.4 Mathematical Machinery 41.4.1 Convex Optimization 41.4.2 Game Theory 61.4.3 “Big Data” Modeled as Large Random Matrices 61.5 Sample Covariance Matrix 101.6 Large Sample Covariance Matrices of Spiked Population Models 111.7 Random Matrices and Noncommutative Random Variables 121.8 Principal Component Analysis 131.9 Generalized Likelihood Ratio Test (GLRT) 131.10 Bregman Divergence for Matrix Nearness 132 Spectrum Sensing: Basic Techniques 152.1 Challenges 152.2 Energy Detection: No Prior Information about Deterministic or Stochastic Signal 152.2.1 Detection in White Noise: Lowpass Case 162.2.2 Time-Domain Representation of the Decision Statistic 192.2.3 Spectral Representation of the Decision Statistic 192.2.4 Detection and False Alarm Probabilities over AWGN Channels 202.2.5 Expansion of Random Process in Orthonormal Series with Uncorrelated Coefficients: The Karhunen-Loeve Expansion 212.3 Spectrum Sensing Exploiting Second-Order Statistics 232.3.1 Signal Detection Formulation 232.3.2 Wide-Sense Stationary Stochastic Process: Continuous-Time 242.3.3 Nonstationary Stochastic Process: Continuous-Time 252.3.4 Spectrum Correlation-Based Spectrum Sensing for WSS Stochastic Signal: Heuristic Approach 292.3.5 Likelihood Ratio Test of Discrete-Time WSS Stochastic Signal 322.3.6 Asymptotic Equivalence between Spectrum Correlation and Likelihood Ratio Test 352.3.7 Likelihood Ratio Test of Continuous-Time Stochastic Signals in Noise: Selin’s Approach 362.4 Statistical Pattern Recognition: Exploiting Prior Information about Signal through Machine Learning 392.4.1 Karhunen-Loeve Decomposition for Continuous-Time Stochastic Signal 392.5 Feature Template Matching 422.6 Cyclostationary Detection 473 Classical Detection 513.1 Formalism of Quantum Information 513.2 Hypothesis Detection for Collaborative Sensing 513.3 Sample Covariance Matrix 553.3.1 The Data Matrix 563.4 Random Matrices with Independent Rows 633.5 The Multivariate Normal Distribution 673.6 Sample Covariance Matrix Estimation and Matrix Compressed Sensing 773.6.1 The Maximum Likelihood Estimation 813.6.2 Likelihood Ratio Test (Wilks Test) for Multisample Hypotheses 833.7 Likelihood Ratio Test 843.7.1 General Gaussian Detection and Estimator-Correlator Structure 843.7.2 Tests with Repeated Observations 903.7.3 Detection Using Sample Covariance Matrices 943.7.4 GLRT for Multiple Random Vectors 953.7.5 Linear Discrimination Functions 973.7.6 Detection of Correlated Structure for Complex Random Vectors 984 Hypothesis Detection of Noncommutative Random Matrices 1014.1 Why Noncommutative Random Matrices? 1014.2 Partial Orders of Covariance Matrices: A < B 1024.3 Partial Ordering of Completely Positive Mappings: (A) < (B) 1044.4 Partial Ordering of Matrices Using Majorization: A ≺ B 1054.5 Partial Ordering of Unitarily Invariant Norms: |||A||| < |||B||| 1094.6 Partial Ordering of Positive Definite Matrices of Many Copies: K k=1 Ak ≤ K k=1 Bk 1094.7 Partial Ordering of Positive Operator Valued Random Variables: Prob(A ≤ X ≤ B) 1104.8 Partial Ordering Using Stochastic Order: A ≤st B 1154.9 Quantum Hypothesis Detection 1154.10 Quantum Hypothesis Testing for Many Copies 1185 Large Random Matrices 1195.1 Large Dimensional Random Matrices: Moment Approach, Stieltjes Transform and Free Probability 1195.2 Spectrum Sensing Using Large Random Matrices 1215.2.1 System Model 1215.2.2 Marchenko-Pastur Law 1245.3 Moment Approach 1295.3.1 Limiting Spectral Distribution 1305.3.2 Limits of Extreme Eigenvalues 1335.3.3 Convergence Rates of Spectral Distributions 1365.3.4 Standard Vector-In, Vector-Out Model 1375.3.5 Generalized Densities 1385.4 Stieltjes Transform 1395.4.1 Basic Theorems 1435.4.2 Large Random Hankel, Markov and Toepltiz Matrices 1495.4.3 Information Plus Noise Model of Random Matrices 1525.4.4 Generalized Likelihood Ratio Test Using Large Random Matrices 1575.4.5 Detection of High-Dimensional Signals in White Noise 1645.4.6 Eigenvalues of (A + B)−1B and Applications 1695.4.7 Canonical Correlation Analysis 1715.4.8 Angles and Distances between Subspaces 1735.4.9 Multivariate Linear Model 1735.4.10 Equality of Covariance Matrices 1745.4.11 Multiple Discriminant Analysis 1745.5 Case Studies and Applications 1755.5.1 Fundamental Example of Using Large Random Matrix 1755.5.2 Stieltjes Transform 1775.5.3 Free Deconvolution 1785.5.4 Optimal Precoding of MIMO Systems 1785.5.5 Marchenko and Pastur’s Probability Distribution 1795.5.6 Convergence and Fluctuations Extreme Eigenvalues 1805.5.7 Information plus Noise Model and Spiked Models 1805.5.8 Hypothesis Testing and Spectrum Sensing 1835.5.9 Energy Estimation in a Wireless Network 1855.5.10 Multisource Power Inference 1875.5.11 Target Detection, Localization, and Reconstruction 1875.5.12 State Estimation and Malignant Attacker in the Smart Grid 1915.5.13 Covariance Matrix Estimation 1935.5.14 Deterministic Equivalents 1975.5.15 Local Failure Detection and Diagnosis 2005.6 Regularized Estimation of Large Covariance Matrices 2005.6.1 Regularized Covariance Estimates 2015.6.2 Banding the Inverse 2035.6.3 Covariance Regularization by Thresholding 2045.6.4 Regularized Sample Covariance Matrices 2065.6.5 Optimal Rates of Convergence for Covariance Matrix Estimation 2085.6.6 Banding Sample Autocovariance Matrices of Stationary Processes 2115.7 Free Probability 2135.7.1 Large Random Matrices and Free Convolution 2185.7.2 Vandermonde Matrices 2215.7.3 Convolution and Deconvolution with Vandermonde Matrices 2295.7.4 Finite Dimensional Statistical Inference 2326 Convex Optimization 2356.1 Linear Programming 2376.2 Quadratic Programming 2386.3 Semidefinite Programming 2396.4 Geometric Programming 2396.5 Lagrange Duality 2416.6 Optimization Algorithm 2426.6.1 Interior Point Methods 2426.6.2 Stochastic Methods 2436.7 Robust Optimization 2446.8 Multiobjective Optimization 2486.9 Optimization for Radio Resource Management 2496.10 Examples and Applications 2506.10.1 Spectral Efficiency for Multiple Input Multiple Output Ultra-Wideband Communication System 2506.10.2 Wideband Waveform Design for Single Input Single Output Communication System with Noncoherent Receiver 2566.10.3 Wideband Waveform Design for Multiple Input Single Output Cognitive Radio 2626.10.4 Wideband Beamforming Design 2686.10.5 Layering as Optimization Decomposition for Cognitive Radio Network 2726.11 Summary 2827 Machine Learning 2837.1 Unsupervised Learning 2887.1.1 Centroid-Based Clustering 2887.1.2 k-Nearest Neighbors 2897.1.3 Principal Component Analysis 2897.1.4 Independent Component Analysis 2907.1.5 Nonnegative Matrix Factorization 2917.1.6 Self-Organizing Map 2927.2 Supervised Learning 2937.2.1 Linear Regression 2937.2.2 Logistic Regression 2947.2.3 Artificial Neural Network 2947.2.4 Decision Tree Learning 2947.2.5 Naive Bayes Classifier 2957.2.6 Support Vector Machines 2957.3 Semisupervised Learning 2987.3.1 Constrained Clustering 2987.3.2 Co-Training 2987.3.3 Graph-Based Methods 2997.4 Transductive Inference 2997.5 Transfer Learning 2997.6 Active Learning 2997.7 Reinforcement Learning 3007.7.1 Q-Learning 3007.7.2 Markov Decision Process 3017.7.3 Partially Observable MDPs 3027.8 Kernel-Based Learning 3037.9 Dimensionality Reduction 3047.9.1 Kernel Principal Component Analysis 3057.9.2 Multidimensional Scaling 3077.9.3 Isomap 3087.9.4 Locally-Linear Embedding 3087.9.5 Laplacian Eigenmaps 3097.9.6 Semidefinite Embedding 3097.10 Ensemble Learning 3117.11 Markov Chain Monte Carlo 3127.12 Filtering Technique 3137.12.1 Kalman Filtering 3147.12.2 Particle Filtering 3187.12.3 Collaborative Filtering 3197.13 Bayesian Network 3207.14 Summary 3218 Agile Transmission Techniques (I): Multiple Input Multiple Output 3238.1 Benefits of MIMO 3238.1.1 Array Gain 3238.1.2 Diversity Gain 3238.1.3 Multiplexing Gain 3248.2 Space Time Coding 3248.2.1 Space Time Block Coding 3258.2.2 Space Time Trellis Coding 3268.2.3 Layered Space Time Coding 3268.3 Multi-User MIMO 3278.3.1 Space-Division Multiple Access 3278.3.2 MIMO Broadcast Channel 3288.3.3 MIMO Multiple Access Channel 3308.3.4 MIMO Interference Channel 3318.4 MIMO Network 3348.5 MIMO Cognitive Radio Network 3368.6 Summary 3379 Agile Transmission Techniques (II): Orthogonal Frequency Division Multiplexing 3399.1 OFDM Implementation 3399.2 Synchronization 3419.3 Channel Estimation 3439.4 Peak Power Problem 3459.5 Adaptive Transmission 3459.6 Spectrum Shaping 3479.7 Orthogonal Frequency Division Multiple Access 3479.8 MIMO OFDM 3499.9 OFDM Cognitive Radio Network 3499.10 Summary 35010 Game Theory 35110.1 Basic Concepts of Games 35110.1.1 Elements of Games 35110.1.2 Nash Equilibrium: Definition and Existence 35210.1.3 Nash Equilibrium: Computation 35410.1.4 Nash Equilibrium: Zero-Sum Games 35510.1.5 Nash Equilibrium: Bayesian Case 35510.1.6 Nash Equilibrium: Stochastic Games 35610.2 Primary User Emulation Attack Games 36010.2.1 PUE Attack 36010.2.2 Two-Player Case: A Strategic-Form Game 36110.2.3 Game in Queuing Dynamics: A Stochastic Game 36210.3 Games in Channel Synchronization 36810.3.1 Background of the Game 36810.3.2 System Model 36810.3.3 Game Formulation 36910.3.4 Bayesian Equilibrium 37010.3.5 Numerical Results 37110.4 Games in Collaborative Spectrum Sensing 37210.4.1 False Report Attack 37310.4.2 Game Formulation 37310.4.3 Elements of Game 37410.4.4 Bayesian Equilibrium 37610.4.5 Numerical Results 37911 Cognitive Radio Network 38111.1 Basic Concepts of Networks 38111.1.1 Network Architecture 38111.1.2 Network Layers 38211.1.3 Cross-Layer Design 38411.1.4 Main Challenges in Cognitive Radio Networks 38411.1.5 Complex Networks 38511.2 Channel Allocation in MAC Layer 38611.2.1 Problem Formulation 38611.2.2 Scheduling Algorithm 38711.2.3 Solution 38911.2.4 Discussion 39011.3 Scheduling in MAC Layer 39111.3.1 Network Model 39111.3.2 Goal of Scheduling 39311.3.3 Scheduling Algorithm 39311.3.4 Performance of the CNC Algorithm 39511.3.5 Distributed Scheduling Algorithm 39611.4 Routing in Network Layer 39611.4.1 Challenges of Routing in Cognitive Radio 39711.4.2 Stationary Routing 39811.4.3 Dynamic Routing 40211.5 Congestion Control in Transport Layer 40411.5.1 Congestion Control in Internet 40411.5.2 Challenges in Cognitive Radio 40511.5.3 TP-CRAHN 40611.5.4 Early Start Scheme 40811.6 Complex Networks in Cognitive Radio 41711.6.1 Brief Introduction to Complex Networks 41811.6.2 Connectivity of Cognitive Radio Networks 42111.6.3 Behavior Propagation in Cognitive Radio Networks 42312 Cognitive Radio Network as Sensors 42712.1 Intrusion Detection by Machine Learning 42912.2 Joint Spectrum Sensing and Localization 42912.3 Distributed Aspect Synthetic Aperture Radar 42912.4 Wireless Tomography 43312.5 Mobile Crowdsensing 43412.6 Integration of 3S 43512.7 The Cyber-Physical System 43512.8 Computing 43612.8.1 Graphics Processor Unit 43712.8.2 Task Distribution and Load Balancing 43712.9 Security and Privacy 43812.10 Summary 438Appendix A Matrix Analysis 441A.1 Vector Spaces and Hilbert Space 441A.2 Transformations 443A.3 Trace 444A.4 Basics of C ∗-Algebra 444A.5 Noncommunicative Matrix-Valued Random Variables 445A.6 Distances and Projections 447A.6.1 Matrix Inequalities 450A.6.2 Partial Ordering of Positive Semidefinite Matrices 451A.6.3 Partial Ordering of Hermitian Matrices 451References 453Index 511

Robert C. Qiu, Department of Electrical and Computer Engineering, Tennessee Technological University, USAProfessor Qiu is currently Director of the Wireless Networking System Laboratory at Tennessee Technological University, USA. He was Founder-CEO and President of Wiscom Technologies, Inc., manufacturing and marketing WCDMA chipsets. Wiscom was acquired by Intel in 2003. Prior to Wiscom, he worked for GTE Labs, Inc. (now Verizon), Waltham, MA, and Bell Labs, Lucent, Whippany, NJ. He holds 5 U.S. patents (another two pending) in WCDMA. He is co-editor of Ultra-Wideband Wireless Communications and Networks (Wiley), and? has authored over 80 technical (journal/conference) papers and contributed 6 book chapters. Professor Qiu has contributed to 3GPP and IEEE standards bodies, and delivered invited seminars to institutions including Princeton University and the U.S. Army Research Lab. Dr. Qiu serves as Associate Editor, IEEE Transaction on Vehicular Technology, International Journal of Sensor Networks and Wireless Communication and Mobile Computing.

Select Guide Rating

• Preface xv1 Introduction 11.1 Vision: “Big Data” 11.2 Cognitive Radio: System Concepts 21.3 Spectrum Sensing Interface and Data Structures 21.4 Mathematical Machinery 41.4.1 Convex Optimization 41.4.2 Game Theory 61.4.3 “Big Data” Modeled as Large Random Matrices 61.5 Sample Covariance Matrix 101.6 Large Sample Covariance Matrices of Spiked Population Models 111.7 Random Matrices and Noncommutative Random Variables 121.8 Principal Component Analysis 131.9 Generalized Likelihood Ratio Test (GLRT) 131.10 Bregman Divergence for Matrix Nearness 132 Spectrum Sensing: Basic Techniques 152.1 Challenges 152.2 Energy Detection: No Prior Information about Deterministic or Stochastic Signal 152.2.1 Detection in White Noise: Lowpass Case 162.2.2 Time-Domain Representation of the Decision Statistic 192.2.3 Spectral Representation of the Decision Statistic 192.2.4 Detection and False Alarm Probabilities over AWGN Channels 202.2.5 Expansion of Random Process in Orthonormal Series with Uncorrelated Coefficients: The Karhunen-Loeve Expansion 212.3 Spectrum Sensing Exploiting Second-Order Statistics 232.3.1 Signal Detection Formulation 232.3.2 Wide-Sense Stationary Stochastic Process: Continuous-Time 242.3.3 Nonstationary Stochastic Process: Continuous-Time 252.3.4 Spectrum Correlation-Based Spectrum Sensing for WSS Stochastic Signal: Heuristic Approach 292.3.5 Likelihood Ratio Test of Discrete-Time WSS Stochastic Signal 322.3.6 Asymptotic Equivalence between Spectrum Correlation and Likelihood Ratio Test 352.3.7 Likelihood Ratio Test of Continuous-Time Stochastic Signals in Noise: Selin’s Approach 362.4 Statistical Pattern Recognition: Exploiting Prior Information about Signal through Machine Learning 392.4.1 Karhunen-Loeve Decomposition for Continuous-Time Stochastic Signal 392.5 Feature Template Matching 422.6 Cyclostationary Detection 473 Classical Detection 513.1 Formalism of Quantum Information 513.2 Hypothesis Detection for Collaborative Sensing 513.3 Sample Covariance Matrix 553.3.1 The Data Matrix 563.4 Random Matrices with Independent Rows 633.5 The Multivariate Normal Distribution 673.6 Sample Covariance Matrix Estimation and Matrix Compressed Sensing 773.6.1 The Maximum Likelihood Estimation 813.6.2 Likelihood Ratio Test (Wilks Test) for Multisample Hypotheses 833.7 Likelihood Ratio Test 843.7.1 General Gaussian Detection and Estimator-Correlator Structure 843.7.2 Tests with Repeated Observations 903.7.3 Detection Using Sample Covariance Matrices 943.7.4 GLRT for Multiple Random Vectors 953.7.5 Linear Discrimination Functions 973.7.6 Detection of Correlated Structure for Complex Random Vectors 984 Hypothesis Detection of Noncommutative Random Matrices 1014.1 Why Noncommutative Random Matrices? 1014.2 Partial Orders of Covariance Matrices: A < B 1024.3 Partial Ordering of Completely Positive Mappings: (A) < (B) 1044.4 Partial Ordering of Matrices Using Majorization: A ≺ B 1054.5 Partial Ordering of Unitarily Invariant Norms: |||A||| < |||B||| 1094.6 Partial Ordering of Positive Definite Matrices of Many Copies: K k=1 Ak ≤ K k=1 Bk 1094.7 Partial Ordering of Positive Operator Valued Random Variables: Prob(A ≤ X ≤ B) 1104.8 Partial Ordering Using Stochastic Order: A ≤st B 1154.9 Quantum Hypothesis Detection 1154.10 Quantum Hypothesis Testing for Many Copies 1185 Large Random Matrices 1195.1 Large Dimensional Random Matrices: Moment Approach, Stieltjes Transform and Free Probability 1195.2 Spectrum Sensing Using Large Random Matrices 1215.2.1 System Model 1215.2.2 Marchenko-Pastur Law 1245.3 Moment Approach 1295.3.1 Limiting Spectral Distribution 1305.3.2 Limits of Extreme Eigenvalues 1335.3.3 Convergence Rates of Spectral Distributions 1365.3.4 Standard Vector-In, Vector-Out Model 1375.3.5 Generalized Densities 1385.4 Stieltjes Transform 1395.4.1 Basic Theorems 1435.4.2 Large Random Hankel, Markov and Toepltiz Matrices 1495.4.3 Information Plus Noise Model of Random Matrices 1525.4.4 Generalized Likelihood Ratio Test Using Large Random Matrices 1575.4.5 Detection of High-Dimensional Signals in White Noise 1645.4.6 Eigenvalues of (A + B)−1B and Applications 1695.4.7 Canonical Correlation Analysis 1715.4.8 Angles and Distances between Subspaces 1735.4.9 Multivariate Linear Model 1735.4.10 Equality of Covariance Matrices 1745.4.11 Multiple Discriminant Analysis 1745.5 Case Studies and Applications 1755.5.1 Fundamental Example of Using Large Random Matrix 1755.5.2 Stieltjes Transform 1775.5.3 Free Deconvolution 1785.5.4 Optimal Precoding of MIMO Systems 1785.5.5 Marchenko and Pastur’s Probability Distribution 1795.5.6 Convergence and Fluctuations Extreme Eigenvalues 1805.5.7 Information plus Noise Model and Spiked Models 1805.5.8 Hypothesis Testing and Spectrum Sensing 1835.5.9 Energy Estimation in a Wireless Network 1855.5.10 Multisource Power Inference 1875.5.11 Target Detection, Localization, and Reconstruction 1875.5.12 State Estimation and Malignant Attacker in the Smart Grid 1915.5.13 Covariance Matrix Estimation 1935.5.14 Deterministic Equivalents 1975.5.15 Local Failure Detection and Diagnosis 2005.6 Regularized Estimation of Large Covariance Matrices 2005.6.1 Regularized Covariance Estimates 2015.6.2 Banding the Inverse 2035.6.3 Covariance Regularization by Thresholding 2045.6.4 Regularized Sample Covariance Matrices 2065.6.5 Optimal Rates of Convergence for Covariance Matrix Estimation 2085.6.6 Banding Sample Autocovariance Matrices of Stationary Processes 2115.7 Free Probability 2135.7.1 Large Random Matrices and Free Convolution 2185.7.2 Vandermonde Matrices 2215.7.3 Convolution and Deconvolution with Vandermonde Matrices 2295.7.4 Finite Dimensional Statistical Inference 2326 Convex Optimization 2356.1 Linear Programming 2376.2 Quadratic Programming 2386.3 Semidefinite Programming 2396.4 Geometric Programming 2396.5 Lagrange Duality 2416.6 Optimization Algorithm 2426.6.1 Interior Point Methods 2426.6.2 Stochastic Methods 2436.7 Robust Optimization 2446.8 Multiobjective Optimization 2486.9 Optimization for Radio Resource Management 2496.10 Examples and Applications 2506.10.1 Spectral Efficiency for Multiple Input Multiple Output Ultra-Wideband Communication System 2506.10.2 Wideband Waveform Design for Single Input Single Output Communication System with Noncoherent Receiver 2566.10.3 Wideband Waveform Design for Multiple Input Single Output Cognitive Radio 2626.10.4 Wideband Beamforming Design 2686.10.5 Layering as Optimization Decomposition for Cognitive Radio Network 2726.11 Summary 2827 Machine Learning 2837.1 Unsupervised Learning 2887.1.1 Centroid-Based Clustering 2887.1.2 k-Nearest Neighbors 2897.1.3 Principal Component Analysis 2897.1.4 Independent Component Analysis 2907.1.5 Nonnegative Matrix Factorization 2917.1.6 Self-Organizing Map 2927.2 Supervised Learning 2937.2.1 Linear Regression 2937.2.2 Logistic Regression 2947.2.3 Artificial Neural Network 2947.2.4 Decision Tree Learning 2947.2.5 Naive Bayes Classifier 2957.2.6 Support Vector Machines 2957.3 Semisupervised Learning 2987.3.1 Constrained Clustering 2987.3.2 Co-Training 2987.3.3 Graph-Based Methods 2997.4 Transductive Inference 2997.5 Transfer Learning 2997.6 Active Learning 2997.7 Reinforcement Learning 3007.7.1 Q-Learning 3007.7.2 Markov Decision Process 3017.7.3 Partially Observable MDPs 3027.8 Kernel-Based Learning 3037.9 Dimensionality Reduction 3047.9.1 Kernel Principal Component Analysis 3057.9.2 Multidimensional Scaling 3077.9.3 Isomap 3087.9.4 Locally-Linear Embedding 3087.9.5 Laplacian Eigenmaps 3097.9.6 Semidefinite Embedding 3097.10 Ensemble Learning 3117.11 Markov Chain Monte Carlo 3127.12 Filtering Technique 3137.12.1 Kalman Filtering 3147.12.2 Particle Filtering 3187.12.3 Collaborative Filtering 3197.13 Bayesian Network 3207.14 Summary 3218 Agile Transmission Techniques (I): Multiple Input Multiple Output 3238.1 Benefits of MIMO 3238.1.1 Array Gain 3238.1.2 Diversity Gain 3238.1.3 Multiplexing Gain 3248.2 Space Time Coding 3248.2.1 Space Time Block Coding 3258.2.2 Space Time Trellis Coding 3268.2.3 Layered Space Time Coding 3268.3 Multi-User MIMO 3278.3.1 Space-Division Multiple Access 3278.3.2 MIMO Broadcast Channel 3288.3.3 MIMO Multiple Access Channel 3308.3.4 MIMO Interference Channel 3318.4 MIMO Network 3348.5 MIMO Cognitive Radio Network 3368.6 Summary 3379 Agile Transmission Techniques (II): Orthogonal Frequency Division Multiplexing 3399.1 OFDM Implementation 3399.2 Synchronization 3419.3 Channel Estimation 3439.4 Peak Power Problem 3459.5 Adaptive Transmission 3459.6 Spectrum Shaping 3479.7 Orthogonal Frequency Division Multiple Access 3479.8 MIMO OFDM 3499.9 OFDM Cognitive Radio Network 3499.10 Summary 35010 Game Theory 35110.1 Basic Concepts of Games 35110.1.1 Elements of Games 35110.1.2 Nash Equilibrium: Definition and Existence 35210.1.3 Nash Equilibrium: Computation 35410.1.4 Nash Equilibrium: Zero-Sum Games 35510.1.5 Nash Equilibrium: Bayesian Case 35510.1.6 Nash Equilibrium: Stochastic Games 35610.2 Primary User Emulation Attack Games 36010.2.1 PUE Attack 36010.2.2 Two-Player Case: A Strategic-Form Game 36110.2.3 Game in Queuing Dynamics: A Stochastic Game 36210.3 Games in Channel Synchronization 36810.3.1 Background of the Game 36810.3.2 System Model 36810.3.3 Game Formulation 36910.3.4 Bayesian Equilibrium 37010.3.5 Numerical Results 37110.4 Games in Collaborative Spectrum Sensing 37210.4.1 False Report Attack 37310.4.2 Game Formulation 37310.4.3 Elements of Game 37410.4.4 Bayesian Equilibrium 37610.4.5 Numerical Results 37911 Cognitive Radio Network 38111.1 Basic Concepts of Networks 38111.1.1 Network Architecture 38111.1.2 Network Layers 38211.1.3 Cross-Layer Design 38411.1.4 Main Challenges in Cognitive Radio Networks 38411.1.5 Complex Networks 38511.2 Channel Allocation in MAC Layer 38611.2.1 Problem Formulation 38611.2.2 Scheduling Algorithm 38711.2.3 Solution 38911.2.4 Discussion 39011.3 Scheduling in MAC Layer 39111.3.1 Network Model 39111.3.2 Goal of Scheduling 39311.3.3 Scheduling Algorithm 39311.3.4 Performance of the CNC Algorithm 39511.3.5 Distributed Scheduling Algorithm 39611.4 Routing in Network Layer 39611.4.1 Challenges of Routing in Cognitive Radio 39711.4.2 Stationary Routing 39811.4.3 Dynamic Routing 40211.5 Congestion Control in Transport Layer 40411.5.1 Congestion Control in Internet 40411.5.2 Challenges in Cognitive Radio 40511.5.3 TP-CRAHN 40611.5.4 Early Start Scheme 40811.6 Complex Networks in Cognitive Radio 41711.6.1 Brief Introduction to Complex Networks 41811.6.2 Connectivity of Cognitive Radio Networks 42111.6.3 Behavior Propagation in Cognitive Radio Networks 42312 Cognitive Radio Network as Sensors 42712.1 Intrusion Detection by Machine Learning 42912.2 Joint Spectrum Sensing and Localization 42912.3 Distributed Aspect Synthetic Aperture Radar 42912.4 Wireless Tomography 43312.5 Mobile Crowdsensing 43412.6 Integration of 3S 43512.7 The Cyber-Physical System 43512.8 Computing 43612.8.1 Graphics Processor Unit 43712.8.2 Task Distribution and Load Balancing 43712.9 Security and Privacy 43812.10 Summary 438Appendix A Matrix Analysis 441A.1 Vector Spaces and Hilbert Space 441A.2 Transformations 443A.3 Trace 444A.4 Basics of C ∗-Algebra 444A.5 Noncommunicative Matrix-Valued Random Variables 445A.6 Distances and Projections 447A.6.1 Matrix Inequalities 450A.6.2 Partial Ordering of Positive Semidefinite Matrices 451A.6.3 Partial Ordering of Hermitian Matrices 451References 453Index 511