Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Early energy estimation in the design process of networked embedded systems

: Heinrich, Patrick; Prehofer, Christian

Fulltext urn:nbn:de:0011-n-2347501 (329 KByte PDF)
MD5 Fingerprint: 2ca14cfca808e4b7c3d86ee6ce94bfd5
Created on: 4.12.2013

Institute for Systems and Technologies of Information, Control and Communication -INSTICC-, Setubal:
PECCS 2013, 3rd International Conference on Pervasive and Embedded Computing and Communication Systems. CD-ROM : Barcelona, Spain, 19 - 21 February 2013, held in conjunction with SENSORNETS 2013 and PHOTOPTICS 2013
Setúbal: INSTICC, 2013
International Conference on Pervasive and Embedded Computing and Communication Systems (PECCS) <3, 2013, Barcelona>
International Conference on Sensor Networks (SENSORNETS) <2, 2013, Barcelona>
International Conference on Photonics, Optics and Laser Technology (PHOTOPTICS) <2013, Barcelona>
Conference Paper, Electronic Publication
Fraunhofer ESK ()
embedded system; energy-efficiency; network-wide optimization; adaptive system; automotive

This paper focuses on estimating energy consumption in networked, embedded systems during the early stages of design. In such systems with long development cycles, early decisions regarding hardware and software have a major impact on energy consumption, which places constraints on the design during later stages of the development process. The main objective of this paper is to model the energy consumption throughout the entire design process and to provide a structure for more detailed models and validation. This requires determining which design decisions and system parameters are available in each of the development phases. Using this approach as a basis, energy consumption models are created for each phase. By relying on different estimation techniques, energy consumption estimates are formulated based on the information that is available at every development phase. The main goal is to illustrate a design flow that produces increasingly accurate estimates in each phase.