Software Defined Radios (eBook)

Sofie Pollin obtained the M.Sc. degree in Electrical Engineering Degree in 2002 and the PhD Degree in Electrical Engineering (with honors) in 2006 from the Katholieke Universiteit Leuven, Belgium. From 2002 to 2006 she was a researcher at the Wireless Research group of the Inter-university Microelectronics Center (IMEC) working on cross-layer energy and performance optimization of wireless systems. From 2005 till 2008 she was a post-doctoral researcher at UC Berkeley working on Cognitive Radio. Currently, she is a senior researcher at IMEC, leading the work on functionality and cognitive behavior of reconfigurable radios. Michael Timmers received the M. Sc. Degree in Electrical Engineering and the Ph.D. degree from the K.U.Leuven, Belgium, in 2005 and 2009, respectively. In September 2005, he joined the TELEMIC division of the Department of Electrical Engineering of K.U.Leuven, where he mainly focused on the application of radio waves in biomedical technology. In March 2006, he joined IMEC to pursue a PhD degree in the field of Cognitive Radio. He was a visiting scholar at the Connectivity Lab at U.C. Berkeley in the summer of 2007. His doctoral thesis focuses on distributed medium access control, Software Defined Radio, Opportunistic Spectrum Access and Cognitive Radio. Liesbet Van der Perre received the M.Sc. degree in Electrical Engineering and the PhD degree from the K.U.Leuven, Belgium, in 1992 and 1997 respectively. Her work in the past focused on radio propagation modelling, system design and digital modems for high-speed wireless communications. She was a system architect in IMECs OFDM ASICs development. Consequently, she was the project leader for IMEC's low power Turbo codec. From 2005 till 2008 she was the scientific director of wireless research group in IMEC and the project leader for the digital baseband Software Defined Radio, and a public speaking coach for IMEC staff. Currently, she is program director for IMEC's green radio program, comprising the cognitive reconfigurable radios and mm-wave communications. Liesbet is a part-time professor at the K.U.Leuven, Belgium, since 2008. She is an author and co-author of over 200 scientific publications published in conference proceedings, journals, and books.

Preface 5
Contents 6
List of Acronyms 9
List of Figures 10
List of Tables 15
Chapter 1: Serving Many Mobile Users in Various Scenarios: Radios to Go Smart(er) and Cognitive 16
1.1 Towards Cognitive Radio 16
1.2 Increasing the Hardware Flexibility 16
1.2.1 Wireless Landscape Giving Challenges and Opportunities 17
1.2.1.1 Heterogeneity Desires Flexibility 17
1.2.1.2 Enabling Seamless Connectivity 18
1.2.1.3 Scaling Technology Imposes Reconfigurability 18
1.2.2 The Software-Defined Radio Solution 19
1.3 Increasing the Policy Flexibility 19
1.3.1 Spectrum: A Scarce Resource 20
1.3.2 The Opportunistic Spectrum Access Solution 20
1.4 Cognitive Radio: Exploiting Flexibility with Intelligent Control 22
1.5 The Need for a New Approach 24
1.6 Radios to Go Smarter and Cognitive 24
Chapter 2: Emerging Standards for Smart Radios: Enabling Tomorrow's Operation 26
2.1 Standards in Evolution 26
2.2 Hardware Flexibility 27
2.2.1 IEEE 802.11: A Flexible Radio Becomes Smarter 28
2.2.1.1 The IEEE 802.11a Physical Layer 28
2.2.1.2 The IEEE 802.11n Physical Layer 29
2.2.1.3 The IEEE 802.11ac Physical Layer 30
2.2.1.4 Multiple Access Through Collision Avoidance with Carrier Sensing 30
2.2.2 3GPP-LTE Evolutions 33
2.2.2.1 The 3GPP-LTE Air Interface 33
2.2.2.2 Multiple Access in 3GPP-LTE 34
2.2.2.3 LTE-Advanced 38
2.3 Spectrum Access Flexibility 38
2.3.1 The ISM Band: Coexistence in Unlicensed Bands 39
2.3.1.1 IEEE 802.11h for Spectrum and Transmit Power Management Extensions 40
2.3.1.2 IEEE 802.19 Coexistence Technical Advisory Group 41
2.3.2 The TV White Spaces: Spectrum Sharing in Licensed Bands 41
2.3.2.1 IEEE 802.22 Wireless Rural Access Networks 43
2.3.2.2 System Overview 44
2.3.2.3 Spectrum Sensing 44
2.4 Operation Across Technologies: Cognitive Radio 46
2.4.1 Mobile Independent Handover: IEEE 802.21 46
2.4.2 Dynamic Spectrum Access Networks: IEEE DYSPAN 47
2.4.2.1 IEEE 1900.4 47
2.4.2.2 1900.6 49
2.4.3 Reconfigurable Radio Systems: ETSI RSS 49
Chapter 3: Cognitive Radio Design and Operation: Mastering the Complexity in a Systematic Way 51
3.1 The Need for a Strategy 51
3.2 The Design Landscape Is No Longer Flat 52
3.3 Design Challenges and Opportunities 53
3.3.1 Design Time Complexity 53
3.3.2 The Mountains We Have to Climb 54
3.3.2.1 Channel Dynamics 54
3.3.2.2 Application Dynamics 55
3.3.2.3 Network Dynamics 55
3.3.3 The Sharing Challenge 56
3.3.3.1 The Spectrum Policy 56
3.3.3.2 Multi-user Interaction 56
3.3.3.3 Fairness 57
3.3.4 Run-Time Complexity 57
3.4 Proposed Control Framework 58
3.4.1 General Design Concepts 58
3.4.1.1 Control Dimensions 58
3.4.1.2 Environment Awareness 59
3.4.1.3 Efficient and Effective Calibration at Run-Time 60
3.4.2 Design-Time Flow 60
3.4.2.1 Design-Time Modeling172 60
3.4.2.2 Identify Control Dimensions173 62
3.4.2.3 Identify Dynamics174 62
3.4.2.4 Mapping to Objective Space175 63
3.4.2.5 Cluster and Monitor System Scenarios176177 63
3.4.2.6 Determine DT Procedure178 64
3.4.3 Run-Time Operation 64
3.4.3.1 Observe the RT Situation172 65
3.4.3.2 Map RT Situation to System Scenario173 65
3.4.3.3 Execute RT Procedure174 65
3.4.3.4 The RT Procedure175 66
3.5 Conclusions 67
Chapter 4: Distributed Monitoring for Opportunistic Radios 68
4.1 To Not Interfere 68
4.1.1 Problem Context 68
4.1.2 Smart Aspect 69
4.1.3 Outdoor 802.11 Measurements 70
4.1.3.1 Measurement Setup 70
4.1.3.2 Observations from Measurements 71
No Clear Trend: 71
Anisotropic Due to Shadowing: 71
Noisy Measurements Due to Fast Fading: 71
4.2 The Sensing Problem 72
4.3 Distributed Distance-to-Contour Estimation 72
4.3.1 Algorithm Overview and Design Decisions 72
4.3.1.1 Local Channel Estimation 73
4.3.1.2 Distance-to-Contour Flooding 73
4.3.1.3 Iterative Power Control 74
4.3.2 Local Channel Estimation 74
4.3.3 Distance-to-Contour Flooding 76
4.3.3.1 Centralized Distance-to-Contour Computation 76
4.3.3.2 Distributed Distance-to-Contour Flooding 77
4.3.4 Iterative Power Control 79
4.3.4.1 The Increasing Power Scenario 79
4.3.4.2 The Decreasing Power Scenario 80
4.3.5 Results 80
4.3.5.1 Simulation Model 80
4.3.5.2 Results and Discussion 82
4.4 Conclusions 83
Chapter 5: Coexistence: The Whole Is Greater than the Sum of Its Parts 85
5.1 Introduction 85
5.2 Modeling Coexistence 86
5.2.1 IEEE 802.15.4 Network Model 86
5.2.2 IEEE 802.11 Interference Model 87
5.2.3 Performance and Energy Measures 88
5.3 Basic Solution: Random Frequency Selection 89
5.4 The Problem from a Different Angle 89
5.5 Scan-Based Approaches 90
5.6 Distributed Learning and Exploration 91
5.6.1 General Framework 91
5.6.2 Learning Engine 92
5.6.3 Exploration Algorithms 92
5.6.3.1 Reward-Based (RB) Exploration 93
5.6.3.2 Finding a Cooling Scheme 93
5.6.3.3 Adaptive Simulated Annealing 94
5.7 Simulation Results 95
5.8 Conclusions 96
Chapter 6: Anticipative Energy and QoS Management: Systematically Improving the User Experience 98
6.1 Energy Efficiency for Smart Radios 98
6.1.1 Minimum Energy at Sufficient QoS 98
6.1.2 Smart Aspects and Energy Efficiency 99
6.2 Anticipation Through Design Time Modeling 100
6.2.1 Flexibility for Energy and QoS 101
6.2.2 The Varying Context 103
6.2.3 Objectives for Efficient Energy and QoS Management 105
6.2.4 Anticipating the Performance 106
6.3 Managing the User Experience 109
6.3.1 Smart Resource Allocation Problem Statement 109
6.3.2 Greedy Resource Allocation 110
6.4 IEEE 802.11a Design Case 112
6.4.1 Energy-Performance Anticipation 113
6.4.2 Anticipative Control in the 802.11 MAC Protocol 115
6.5 Adapting to the Dynamic Context 117
6.6 Conclusions 118
Chapter 7: Distributed Optimization of Local Area Networks 120
7.1 Introduction 120
7.2 Existing Flexibility and Control Mechanisms 121
7.2.1 Optimization of IEEE 802.11 Networks 121
7.2.1.1 Transmission Rate 121
7.2.1.2 Carrier Sense Threshold 122
7.2.1.3 Transmit Power 122
7.2.1.4 Hybrid Control 123
7.2.2 Benchmark Solution: Spatial Backoff 123
7.2.2.1 Algorithm Overview 123
7.2.2.2 Opportunities 124
7.2.3 Multi-Agent Learning 124
7.3 Spatial Learning: Distributed Optimization of IEEE 802.11 Networks 125
7.3.1 The General Framework 125
7.3.2 The Control Dimensions 127
7.3.2.1 Rate 127
7.3.2.2 Transmission Power 127
7.3.2.3 Carrier Sense Threshold 127
7.3.3 System Scenarios 128
7.3.4 Design-Time Procedures 129
7.3.5 The Learning Engine 131
7.3.6 Seeding the Learning Engine with the DT Procedures 132
7.3.7 Implementation in the IEEE 802.11 MAC Protocol 133
7.3.7.1 The Reward 133
7.3.7.2 Theoretical Throughput Estimation 134
7.3.7.3 Observation Reuse 134
7.4 Assessing the Gains 135
7.5 Conclusions 139
Chapter 8: Close 140
8.1 "Good Enough" Is "Close Enough to Optimal" 140
8.2 Closing Remarks: The End Is Not There nor in Sight 142
8.2.1 Keep Moving with the Target 142
References 144