Inhalt

•  
   Abstract
•  
   Zusammenfassung
•  
   Contents
•  
  1 Introduction
•  
   1.1 Context in Avionics
•  
   1.2 Problem Statement
•  
   1.3 Contributions & Document Structure
•  
   2 Basic Concepts
•  
  2.1 Dependability
•  
   2.1.1 Availability
•  
   2.1.2 Reliability
•  
   2.1.3 Safety
•  
   2.1.4 Integrity
•  
   2.2 Certification
•  
   2.2.1 Development Assurance Levels
•  
  2.2.2 Safety Analysis
•  
   2.2.2.1 Qualitative Design Analysis
•  
   2.2.2.2 Quantitative Design Analysis
•  
   2.2.3 Mixed Criticality Certification
•  
   2.2.4 Commercial off-the-shelf Devices
•  
   2.3 Real-Time Communication Systems
•  
  2.3.1 Mechanisms for Bounded Latency
•  
   2.3.1.1 Collision Detection
•  
   2.3.1.2 Time-Based Mechanisms
•  
   2.3.1.3 Token-Based & Master-Slave Mechanisms
•  
   2.3.1.4 Switched Mechanisms
•  
   2.3.2 Heterogeneous Latency Requirements
•  
   2.4 Verification of Performance Bounds
•  
   2.4.1 Network Calculus
•  
   2.4.2 Model Checking with Timed-Automata
•  
   2.4.3 Trajectory Approach
•  
   2.5 Summary
•  
   3 Related Work
•  
  3.1 Requirements and Challenges
•  
   3.1.1 Real-Time
•  
  3.1.2 Fault Tolerance
•  
   3.1.2.1 Fault Assumptions
•  
   3.1.3 Self-Configuration
•  
   3.2 Aeronautical Networks
•  
  3.2.1 Strictly Aeronautical Networks
•  
   3.2.1.1 Avionics Full-Duplex Switched Ethernet
•  
   3.2.1.2 Time Triggered Protocol
•  
   3.2.1.3 Legacy Networks
•  
  3.2.2 Adopted Automotive & Industrial Networks
•  
   3.2.2.1 IEC High-Availability Seamless Redundancy
•  
   3.2.2.2 IEEE AVB and TSN
•  
   3.2.2.3 OpenFlow
•  
   3.2.3 Conclusion with Research Gap
•  
  3.3 Self-Configuration
•  
   3.3.1 Management Data & Configuration Channels
•  
   3.3.2 Topology Discovery
•  
   3.3.3 Active Topology Control
•  
   3.3.4 Stream Reservation
•  
   3.3.5 Self-Configuration
•  
   3.3.6 Conclusion with Research Gap
•  
   4 Future Avionic Networks
•  
  4.1 Priority-Based Forwarding in Switches with a Global Timebase
•  
   4.1.1 Definitions
•  
   4.1.2 Mathematical Model
•  
   4.1.2.1 One TAQ per Node
•  
   4.1.2.2 Multiple TAQs per Node
•  
   4.1.2.3 Specifics to High Availability Seamless Redundancy
•  
   4.1.2.4 Gain of Time-Aware Queues
•  
   4.1.3 Simulation of Extended HSR
•  
   4.1.4 Comparison of Mathematical Model and Simulation Results
•  
   4.1.4.1 Normal Working Mode without Failure
•  
   4.1.4.2 Single Link Failure Case
•  
   4.1.4.3 Loss of Time Synchronization
•  
  4.1.5 Conclusion
•  
   4.1.5.1 Open Problems and Future Work
•  
  4.2 Real-Time OpenFlow Network with Non-Synchronized Switches
•  
  4.2.1 Definition & Concept
•  
   4.2.1.1 Token Bucket in AFDX and OpenFlow
•  
   4.2.1.2 Deterministic Behavior and End-to-End Latencies
•  
   4.2.1.3 Matching of Virtual Links
•  
   4.2.1.4 Role of the OpenFlow Controller
•  
   4.2.2 Experiments and Results
•  
   4.2.2.1 Hardware vs. Software Matching
•  
   4.2.2.2 Meter Accuracy
•  
   4.2.2.3 Multiple Metered Streams
•  
   4.2.2.4 Failing End-System
•  
   4.2.3 Real-World Scenario
•  
   4.2.4 Conclusion
•  
  4.3 IEEE Time Sensitive Networking (TSN) with Fault-Tolerance and Frame Preemption
•  
   4.3.1 Overview of IEEE Time Sensitive Networking (TSN)
•  
   4.3.1.1 Explicit Stream Identification
•  
   4.3.1.2 Per-Stream Filtering and Policing (PSFP)
•  
   4.3.1.3 Frame Replication and Elimination for Reliability (FRER)
•  
   4.3.1.4 Frame Premption (Qbu)
•  
   4.3.2 TSimNet Simulation Framework
•  
   4.3.2.1 Non-Standard Additions
•  
   4.3.2.2 Computational Overhead
•  
   4.3.3 Evaluation
•  
   4.3.3.1 Simple Line Congestion
•  
   4.3.3.2 Industrial Line-Topology Preemption
•  
   4.3.3.3 Industrial TSN Profile
•  
   4.3.4 Conclusion
•  
  4.4 Results & Evaluation
•  
   4.4.1 Evaluation
•  
   4.4.2 Interpretation
•  
   5 Self-Configuring Network
•  
  5.1 System Model
•  
   5.1.1 Configurable Parameters
•  
  5.2 Self-Configuring Plug and Play Real-Time Network
•  
  5.2.1 Concept & System Description
•  
   5.2.1.1 Network Control and Topology Awareness
•  
   5.2.1.2 Service Discovery
•  
   5.2.1.3 Constraining Traffic
•  
   5.2.1.4 Reaction to Failures
•  
   5.2.2 Implementation
•  
   5.2.3 Discussion & Evaluation
•  
   5.2.4 Conclusion
•  
   5.3 A Network with In-band Configuration
•  
  5.3.1 Concept
•  
   5.3.1.1 Bootstrap Process
•  
   5.3.1.2 Fail-Over
•  
   5.3.1.3 Determinism of Safe Control Channel
•  
   5.3.2 Model
•  
   5.3.3 Implementation
•  
   5.3.3.1 Configure Policing
•  
   5.3.3.2 Topology Discovery & Route Placement
•  
   5.3.4 Evaluation
•  
   5.3.4.1 Bootstrap Times
•  
   5.3.4.2 Fail-Over
•  
   5.3.4.3 Determinism
•  
   5.3.5 Conclusion
•  
  5.4 Results & Evaluation
•  
   5.4.1 Evaluation
•  
   5.4.2 Interpretation
•  
  6 Conclusion & Future Work
•  
   6.1 Summary & Conclusion
•  
   6.2 Future Research Directions
•  
   A Author's Contribution
•  
   B Bibliography