Title :
Fieldbus scheduling using Boltzmann machine
Author :
AIi, Usama M ; Al Jebory, Karim
Author_Institution :
Tokyo Electr. Power services CO., Ltd., Tokyo, Japan
Abstract :
A centralized access control in fieldbus systems is characterized by the presence of a processing unit acting as link active scheduler (LAS), whose task is to manage the bandwidth, distributing it among all the producing devices, and respecting their time constraints. The bandwidth is distributed on the basis of a scheduling table containing transmission instants for the information produced by different processes in the system such as to guarantee correct scheduling. This paper discusses the results of using Boltzmann machine neural networks (NNs) with reduced scheduling tables. The work is proposed to replace the existing systems in many remote contra stations connecting intelligent field devices.
Keywords :
Boltzmann machines; bandwidth allocation; field buses; scheduling; Boltzmann machine neural network; bandwidth management; centralized access control; fieldbus scheduling; fieldbus systems; intelligent field devices; link active scheduler; reduced scheduling tables; remote contra station; Neurons; Processor scheduling; Factory communication; Field Bus; Machine Learning and AI; Neural Networks; Real Time Decision Algorithms; Real Time Systems;
Conference_Titel :
Systems Signals and Devices (SSD), 2010 7th International Multi-Conference on
Conference_Location :
Amman
Print_ISBN :
978-1-4244-7532-2
DOI :
10.1109/SSD.2010.5585548