Genetic algorithm based ARINC-664 worst case delay optimization using network calculus

Abstract

The Integrated Modular Avionics (IMA) concept, which enables dynamic resource sharing to provide challenging requirements of next-generation aircraft, has been gained significant attention. Deterministic network (DTN) is a key technology of the IMA platform to provide communication among sub-systems with bounded latency and jitter. The quantity of data transferred across IMA systems becomes more essential as the number of integrated functions within an aircraft grows. Because some of the existing avionics communication protocols cannot keep up with the increase in the number of communicating nodes, Ethernet-based DTN solutions have been proposed. Ethernet-based solutions such as ARINC-664, IEEE Time-triggered Ethernet (TTEthernet), and Time Sensitive Network (TSN) are commonly used for DTN. ARINC-664 is the most widely used solution since it does not require a complex synchronization mechanism, and hence, its certification is relatively easier. Ethernet-based deterministic network solution requires offline network planning to make sure that strict timing requirements are met. Network Calculus (NC) is heavily used tool to calculate the worst-case delay bounds for each ARINC-664 network configuration. This thesis presents a Genetic Algorithm (GA) based ARINC-664 network delay optimization using the NC, where the GA is used to effectively search the mapping of ARINC-664 Virtual Links (VLs) to priority levels using the extended priority scheme. While there are only two priority levels in the ARINC-664 standard, the extended priority concept increases the number of priority levels to improve the schedulability of VLs. For each possible assignment of the VLs to the priority levels, the NC analysis provides the worst-case delay results for all VLs. We define three different fitness functions aiming to minimize the maximum, the average, and the standard deviation of the worst-case VL delays, respectively. The results demonstrate that the extended priority concept improves the schedulability of VLs and the GA optimization approach can successfully achieve the desired objectives for the VL delays if the appropriate cost function is selected. In this thesis, we also implement the ARINC-664 End System (ES) and Switch (SW) simulation model in OMNeT++ which is used to simulate and analyze computer networks to verify the NC results of the extended priority concept. The simulation results are always lower than the NC bounds, indicating that the proposed concept successfully provides delay optimization.