Enhancing Embedded Systems Simulation : A Chip-Hardware-in-the-Loop Simulation Framework

Using the real Microcontroller (μC) as a replacement for a μC model inside a system simulation of a μC-based system is a big benefit since the μC is already verified. Moreover, its maximum performance and its accuracy are much higher than any simulation model. With the Chip-Hardware-in-the-Loop Simulation (CHILS) approach, Christian Köhler covers the connection between μC and simulation, the analysis and optimization of such coupling systems as well as the interface abstraction. He develops the hardware to simulation coupling system with a focus on less hardware effort, the capabilities to couple with different simulation environments, and the efficiency of coupling. The interface abstraction primarily supports the efficiency of coupling. The system analysis and optimization concepts include formal criteria to determine the fidelity of Hardware-in-the-Loop (HIL) coupling systems, stability analysis of coupling systems, and the numerical analysis of applied software algorithms. Furthermore, the author presents existing concepts to simulate complex systems and compares them with the new approach. Cover......Page 1 Enhancing Embedded Systems Simulation ......Page 2 Acknowledgment......Page 6 Contents......Page 8 List of Figures......Page 12 List of Tables......Page 16 Acronyms......Page 18 1.1 Motivation......Page 24 1.2 Objective of the Thesis......Page 25 1.3 Thesis Organization......Page 26 2.1 Work Hypothesis......Page 28 2.2 Basics......Page 31 2.2.1 Motivation for Modelling and Simulation......Page 32 2.2.2 Modelling Basics......Page 34 2.2.3 Simulation Systems......Page 35 2.2.3.1 Continuous Simulation......Page 36 2.2.3.2 Discrete Simulation......Page 37 2.3 Possibilities of Hardware-Simulation-Coupling......Page 38 2.3.1 Motivation for Hardware-in-the-Loop......Page 39 2.3.2 Commercial Hardware-in-the-Loop Solutions......Page 40 2.3.2.2 dSPACE......Page 41 2.3.2.4 National Instruments......Page 42 2.3.3.1 University of South Carolina – Virtual Test Bed......Page 43 2.3.4 Coupling Concepts......Page 44 2.4.2 Software-in-the-Loop Simulation......Page 46 2.4.3 Hardware-in-the-Loop Simulation......Page 47 2.4.4 Emulation and Emulator......Page 48 2.6 Subproblems of Hardware-Simulation-Coupling......Page 49 2.6.3 Event Exchange Optimization......Page 50 3.1.1 Simulator Coupling and Co-Simulation......Page 52 3.1.2 Synchronization of Co-Simulations......Page 54 3.2 Related Work......Page 55 3.3 Exchange of Events between Microcontroller and Simulation......Page 58 3.3.1.1 Basic Coupling Principles......Page 59 3.3.2 Event Discrete Simulation......Page 61 3.3.2.1 Basic Coupling Principles......Page 62 3.3.3 Conclusions......Page 64 3.4 CHILS Event Exchange Mechanism......Page 65 3.5.1 Measurement of the Timing Difference......Page 68 3.5.1.2 GPTA......Page 69 3.5.2.1 Additional Instructions......Page 71 3.5.2.3 Pipelining......Page 73 3.5.3 Possibilities for Compensation......Page 74 3.6 Summary......Page 76 4.1 Related Work......Page 78 4.2 Interface Modelling and Abstraction......Page 81 4.2.2 Interface Representation......Page 82 4.2.2.2 Analogue Interfaces......Page 84 4.2.2.3 Communication Interfaces......Page 85 4.3 Summary......Page 87 5 Optimization – Coupling System Analysis......Page 88 5.1 Related Work......Page 89 5.2 Basics......Page 90 5.3.1 Transparency and Fidelity Deffnition......Page 91 5.4 Example......Page 97 5.5 Comparison of Di.erent Coupling Systems......Page 100 5.5.2 Interpretation of Data Sheets......Page 101 5.5.3 Scenarios......Page 103 5.5.4.1 DeskPODTM Hardware......Page 104 5.5.5.1 dSPACE Hardware......Page 105 5.5.6 Comparison Results......Page 106 5.6 Summary......Page 108 6.1 Analysis of Algorithms......Page 110 6.1.1 Related Works......Page 111 6.1.2.1 Error Propagation......Page 112 6.1.3 Arithmetic Basic Operations......Page 113 6.1.3.1 Combination of Arithmetic Basic Operations......Page 114 6.1.4 Algorithm Analysis Process......Page 115 6.1.4.1 CFG-DFG Analysis......Page 116 6.1.4.2 Training Set......Page 120 6.2 Classi.cation of Algorithms......Page 121 6.2.1 Related Work......Page 122 6.2.2.1 Graph Matching......Page 123 6.2.2.2 Recombination......Page 125 6.2.2.3 Experimental Results......Page 126 6.3 Analysis and Classi.cation of Systems......Page 127 6.3.1 Stability of Linear Time Invariant Systems......Page 129 6.3.2 Stability of Nonlinear Time Invariant Systems......Page 130 6.4.1 Stability of Control Loops......Page 132 6.4.2 Example......Page 133 6.4.5 Control Loop Analysis......Page 134 6.4.6 Control Analysis – Numerical......Page 135 6.5 Summary......Page 137 7.1.1 Events......Page 140 7.2.1 Execution Time Accuracy......Page 141 7.2.2 Event Distribution Accuracy......Page 142 7.2.3 Event Occurrence Accuracy......Page 144 7.3 Example......Page 145 7.4 Summary......Page 147 8.1.1 Data Exchange and Synchronization......Page 150 8.1.2 Interface Abstraction......Page 152 8.2.1.1 Microcontroller Peripherals......Page 153 8.2.1.2 Monitor Context......Page 154 8.3.2 Rapid Control Prototyping......Page 155 8.3.3 Test Applications......Page 156 8.4.1 Simulation Coupling – CHILS-API......Page 157 8.4.4.1 Coupling System Analysis......Page 159 8.4.4.2 Algorithm Analysis......Page 160 8.4.4.3 CHILS Runtime Pro.ling......Page 161 8.5 Optimization Flow......Page 162 8.6 Summary......Page 164 9.1 Software Models on Di.erent Levels of Abstraction......Page 166 9.1.2 Instruction Accurate Simulation......Page 168 9.1.5 Register Transfer Level......Page 169 9.2.1.1 Pin-Based Connection......Page 170 9.2.1.2 Embedding via Communication Interface ......Page 172 9.2.2 Emulation......Page 174 9.3 Comparison......Page 175 9.4.2 CHILS Performance......Page 182 9.5 Summary......Page 185 10.1 SystemC-Coupling......Page 186 10.1.1 Virtual Peripheral Extension......Page 189 10.2 Matlab/Simulink-Coupling......Page 191 10.2.1 CHILS-Demonstration Platform......Page 193 10.2.1.1 Results......Page 194 11.1.1 Connection between Hardware and Simulation......Page 198 11.1.3 Event Exchange Optimization......Page 199 11.1.4 CHILS Framework......Page 200 11.2.1 Driver Support......Page 201 11.2.3 Microcontroller Extensions......Page 202 11.2.6 Algorithm Analysis......Page 203 A.1.1 Fidelity Functions of System Comparison......Page 204 B.1.1 Time Di.erence Measurement......Page 206 B.1.2 Time Di.erence Measurement – dsync/isync......Page 207 B.2.1 Result Tables of System Comparison......Page 208 B.3.1 Synthetic Simulation Performance......Page 210 C.1.1 CHILS-API......Page 212 C.2.1 MATLAB®/Simulink® s-function......Page 215 C.2.2 SystemC......Page 220 D.1 Chapter 2......Page 224 D.2.2 dSpace Datasheets......Page 225 Bibliography......Page 228 Glossary......Page 240 Annotation Christian Köhler covers the connection between C and simulation, the interface abstraction as well as the analysis and optimization of coupling systems with the Chip-Hardware-in-the-Loop Simulation (CHILS) approach. He develops the hardware to simulation coupling system with a focus on less hardware effort, the capabilities to couple with different simulation environments, and the efficiency of coupling. Furthermore, the author presents existing concepts to simulate complex systems and compares them with the new approach Christian Köhler covers the connection between μC and simulation, the interface abstraction as well as the analysis and optimization of coupling systems with the Chip-Hardware-in-the-Loop Simulation (CHILS) approach. He develops the hardware to simulation coupling system with a focus on less hardware effort, the capabilities to couple with different simulation environments, and the efficiency of coupling. Furthermore, the author presents existing concepts to simulate complex systems and compares them with the new approach. Christian Koehler covers the connection between C and simulation, the interface abstraction as well as the analysis and optimization of coupling systems with the Chip-Hardware-in-the-Loop Simulation (CHILS) approach.