ABSTRACT
For most real-time signal processing applications there are many ways to realize them in terms of a specific algorithm. As reported by system designers, in practice this choice is mainly based on “cost” measures such as the number of components, performance, pin count, power consumption, and the area of the custom components. Currently, due to design time restrictions, the system designer has to select – on an ad-hoc basis – a single promising path in the huge decision tree from abstract specification to more refined specification (Fig. 26.1). To alleviate this situation, there is a need for fast and early feedback at the algorithm level without going all the way to assembly code or hardware layout. Only when the design space has been sufficiently explored at a high level and when a limited number of promising candidates have been identified, a more thorough and accurate evaluation is required for the final choice (Fig. 26.1). System exploration environment: envisioned situation. https://s3-euw1-ap-pe-df-pch-content-public-p.s3.eu-west-1.amazonaws.com/9781315214719/0fb871de-cdc7-4110-bdc0-1b5d88b350ec/content/fig26_1.tif"/>
