System-Level Power and Energy Estimation Methodology for Open Multimedia Applications Platforms

In this paper, we propose a power and energy estimation methodology at system-level for Open Multimedia Applications Platforms (OMAPs). Within this methodology, the Functional-Level Power Analysis (FLPA) is extended to create generic power models of the platform under test. Then, a simulation framework is developed at the transactional-level to accurately evaluate the activities used in the previously developed power and energy models. The proposed methodology has several benefits: it considers the power and energy consumption of the entire platform including peripherals and leads to accurate estimates. The efficiency of the proposed system-level methodology is validated using mono-processor and heterogeneous multiprocessor embedded architectures designed around OMAPs. The estimated power and energy results provide a maximum error of 5% for mono-processor and 9% for heterogeneous multiprocessor based system when compared against the real board measurements.