Genetic Algorithm for Scheduling Communication Networks in Time-Triggered Systems-of-Systems

Abstract

Systems engineering and management have evolved from developing distributed systems to the integration of complex adaptive systems and the advent of Systems of Systems (SoS). The predictable collaboration among constituent systems in an SoS plays a crucial role, especially in time-critical applications. In an SoS, we face independent constituent systems without global knowledge and central control. At the same time, resource reservations using appropriate scheduling algorithms are required to satisfy real-time requirements. Most of the existing scheduling solutions are defined for monolithic systems or complex systems with centralized authorities. In this paper, we present Time-Triggered Systems of Systems (TTSoS) and introduce a scheduling model for SoS applications with real-time requirements. We propose a two-level iterative genetic algorithm (GA) for scheduling the tasks and messages within and between time-triggered constituent systems. The results show the capability of temporal guarantees. Also, the results of the GA scheduler are compared with the performance of local search heuristics for the generated examples.