Hierarchical neural networks based prediction and control of dynamic reconfiguration for multilevel embedded systems

Authors:
Akram Eddahech;Sofien Chtourou;Mohamed Chtourou
Affiliations:
Ecole Nationale d'Ingénieurs de Sfax, BP 3038, Sfax, Tunisia;Institut supérieur d'informatique et de multimédia de Sfax, BP 242 Sfax 3021, Tunisia;Ecole Nationale d'Ingénieurs de Sfax, BP 3038, Sfax, Tunisia
Venue:
Journal of Systems Architecture: the EUROMICRO Journal
Year:
2013

Citing 18
Cited 0

A simulation study of artificial neural networks for nonlinear time-series forecasting

Computers and Operations Research
Performance of reconfigurable architectures for image-processing applications

Journal of Systems Architecture: the EUROMICRO Journal - Special issue: Reconfigurable systems
Creating an adaptive embedded system by applying multi-agent techniques to reconfigurable hardware

Future Generation Computer Systems - Special issue: Computational science of lattice Boltzmann modelling
A Design Methodology for Dynamic Reconfiguration: The Caronte Architecture

IPDPS '05 Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Workshop 3 - Volume 04
A new Multilevel Hierarchical MFPGA and its suitable configuration tools

ISVLSI '06 Proceedings of the IEEE Computer Society Annual Symposium on Emerging VLSI Technologies and Architectures
Advanced hardware/software co-design on reconfigurable network-on-chip based hyper-platforms

Computers and Electrical Engineering
Forecasting time series with a new architecture for polynomial artificial neural network

Applied Soft Computing
Applying neural networks to performance estimation of embedded software

Journal of Systems Architecture: the EUROMICRO Journal
A monitoring-aware network-on-chip design flow

Journal of Systems Architecture: the EUROMICRO Journal
A Self-Reconfigurable Platform for Scalable DCT Computation Using Compressed Partial Bitstreams and BlockRAM Prefetching

ISVLSI '09 Proceedings of the 2009 IEEE Computer Society Annual Symposium on VLSI
Dynamic Reconfiguration of Two-Level Caches in Soft Real-Time Embedded Systems

ISVLSI '09 Proceedings of the 2009 IEEE Computer Society Annual Symposium on VLSI
Time-series prediction using a local linear wavelet neural network

Neurocomputing
Multi-level reconfigurable architectures in the switch model

Journal of Systems Architecture: the EUROMICRO Journal
Online adaptive utilization control for real-time embedded multiprocessor systems

Journal of Systems Architecture: the EUROMICRO Journal
Architectures and optimization methods of flash memory based storage systems

Journal of Systems Architecture: the EUROMICRO Journal
Data management for component-based embedded real-time systems: The database proxy approach

Journal of Systems and Software
Reconfigurable embedded systems: an application-oriented perspective on architectures and design techniques

SAMOS'05 Proceedings of the 5th international conference on Embedded Computer Systems: architectures, Modeling, and Simulation
Expressing embedded systems configurations at high abstraction levels with UML MARTE profile: Advantages, limitations and alternatives

Journal of Systems Architecture: the EUROMICRO Journal

Quantified Score

Hi-index	0.00

Visualization

Abstract

Multimedia design such as video decoders are typically composed of several communicating tasks. Each task is characterized by its workload variation. The target device of this kind of application contains several processing unit. This calls for a dynamic management of hardware units to improve the QOS of the application and to optimally allocate resources. In this paper, we propose a new architecture based on hierarchical multilevel neural network to model workload variation of each task. The hierarchical structure of this neural network perfectly describes the multilevel decomposition of each hardware unit. The aim of this investigation is to build a design with a control unit that manages the architecture and resource allocation according to the neural network workload prediction.