Title :
A Neural Network Approach to Dynamic Frequency Scaling
Author :
Lahiri, Anirban ; Bussa, Nagaraju ; Saraswat, Pawan
Abstract :
An intelligent dynamic frequency scaling strategy has been proposed in this paper. The technique uses a neural network model to predict the future computational load of the system and uses this information for taking decisions related to frequency scaling. The mentioned policy is highly effective for applications with complex behavior and drastically varying computational load.
Keywords :
embedded systems; mobile computing; neural nets; intelligent dynamic frequency scaling strategy; mobile embedded systems; neural network approach; portable embedded systems; Clocks; Computer networks; Dynamic voltage scaling; Embedded system; Frequency; History; MPEG 4 Standard; Neural networks; Predictive models; Switches;
Conference_Titel :
Advanced Computing and Communications, 2007. ADCOM 2007. International Conference on
Conference_Location :
Guwahati, Assam
Print_ISBN :
0-7695-3059-1
DOI :
10.1109/ADCOM.2007.123
