Bachelorprojekt - Softwareteknologi | Project No. 0248: Routing and scheduling for TSN Networks |
Aktuelle | Tidligere |
Networking (TSN) is a set of amendments that extend Ethernet to support distributed safety-critical and real-time applications in the industrial automation, aerospace and automotive areas. TSN is intended for real-time and safety-critical applications that have to meet stringent timing requirements and tolerate failures. You can read more about TSN via this OneDrive link (use your DTU login). For this project you will have to implement a solution to the combinatorial optimization problem of routing and scheduling streams using TSN. See this OneDrive link with lectures from DTU’s Systems Optimization course to learn more about combinatorial optimization and solutions such as metaheuristics (Simulated Annealing, Genetic Algorithms) and mathematical programming (e.g., Constraint Programming).
As an input to the optimization problem you have an architecture model consisting of a TSN network topology and an application model consisting of a set of streams. For example, the architecture could model an in-vehicle TSN-based platform and the application could model a semi-autonomous driving application; or the architecture could be the backbone of a smart factory and the applications could be related to industrial control. The architecture is modeled as a directed graph of a TSN network. A vertex in this graph can either be an end system (sender/receiver of streams) or a switch. The edges of the graph represent directional network links with a given speed. The applications are modeled as a set of periodic hard real-time streams. For each stream we know the sender and receiver end systems in the TSN network graph, the size in Bytes, period and deadline. You have to implement a software program that determines the routing and scheduling of all given streams using any optimization approach (e.g., metaheuristics, Constraint Programming. You may use any programming language and libraries.
Supervisor(s) Paul Pop
Sidst opdateret: Nov 18, 2020 af Hans Henrik Løvengreen |