Design and evaluation of high-level estimation and optimization techniques for power consumption
This report presents design and evaluation of High-Level Estimation and Optimization Techniques for Power Consumption. Power dissipation has become increasingly more important as a design constraint as factors such as growing portable devices market, reliability, cooling and packaging...
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Final Year Project |
| Language: | English |
| Published: |
2009
|
| Subjects: | |
| Online Access: | http://hdl.handle.net/10356/17879 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Institution: | Nanyang Technological University |
| Language: | English |
| Summary: | This report presents design and evaluation of High-Level Estimation and Optimization
Techniques for Power Consumption.
Power dissipation has become increasingly more important as a design constraint as
factors such as growing portable devices market, reliability, cooling and packaging are
given a heavier consideration when designing ICs. By obtaining estimation of power
consumption at higher levels of abstraction, designers are able to optimize and reduce
power consumption more significantly and thus reducing design iteration time.
In this project, we investigated the characterization based macro modeling technique
by building a power model using information of lower level implementations.
Replicating the steps taken, we have determined the inputs used are in normal binary
format rather than 2’s complement binary. At the same time, the steps taken were
automated whenever possible. Finally, the model was verified against the results from
Synopsys Nanosim and an accuracy of 8.07% was obtained. |
|---|