Neural scheduler for real-time networks

Scheduling problem for real-time systems is known to be intractable, i.e. NP-hard or completely NP for most cases. A neural computation technique is introduced to solve it within a limited time under a hard real-time environment. Although the neural computation can be effectively carried out by a mu...

Full description

Saved in:
Bibliographic Details
Main Authors: Ae, T., Yamashita, M., Nitta, K.-I.
Format: Conference Proceeding
Language:English
Subjects:
Computer networks
Intelligent networks
Neural networks
Polynomials
Process control
Process design
Processor scheduling
Real time systems
Scheduling algorithm
Telecommunication computing
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Scheduling problem for real-time systems is known to be intractable, i.e. NP-hard or completely NP for most cases. A neural computation technique is introduced to solve it within a limited time under a hard real-time environment. Although the neural computation can be effectively carried out by a multiprocessor, a special processor is designed to obtain the scheduled result with no overhead.< >
DOI:10.1109/HICSS.1989.48071