A New Behavioural Power Modelling Approach for FPGA based Custom Cores

Abstract

Field Programmable Gate Arrays (FPGAs) are gaining acceptance as the preferred platform for a number of applications including those with mobility and performance constraints. While FPGAs continue to scale impressively in terms of device geometries, multi-million gate capacities add near-ASIC performance, newer issues such as power consumption and energy aware design seek urgent attention. While FPGA design tools have reached considerable levels of maturity in a short span of time, a lot of work remains to be done in the area of high level power estimation which needs to be integrated into the design cycle. In this paper, a novel behavioural/functional level power modelling methodology called " Functional Level Power Analysis and Modelling (FLPAM)" that provides a good trade off between complexity and accuracy, and enables the designer to achieve incremental improvements in power and energy metrics throughout the design process has been presented. The proposed approach is successfully validated using custom IP cores for FPGAs on a number platforms.