Design space exploration and synthesis for digital signal processing algorithms from Simulink models

Abstract

Design teams are increasingly looking for design flows that can rapidly lead to high performance and low power implementation of DSP algorithms. Model-based design can satisfy this requirement, but it must be (1) coupled with efficient high-level synthesis support in order to provide good Quality of Results, and (2) controlled to derive the desired area/performance/throughput trade-off. We present a semi-automatic design flow for rapid high level synthesis-based hardware design space exploration starting from Simulink digital signal processing models. We illustrate our flow with a realistically complex signal processing algorithm for estimating the direction of arrival of a sound source. We show how one can start from a functionally validated fixed point model in Simulink and then go through a relatively simple design flow for hardware synthesis and automatic design space exploration, obtaining a very efficient hardware implementation that is competitive with the RTL implementation generated by another commercial model-based design tool.