Static scheduling of synchronous data flow programs for digital signal processing
IEEE Transactions on Computers
Statecharts: A visual formalism for complex systems
Science of Computer Programming
Comparing models of computation
Proceedings of the 1996 IEEE/ACM international conference on Computer-aided design
YAPI: application modeling for signal processing systems
Proceedings of the 37th Annual Design Automation Conference
The development of a high-level synthesis system for concurrent vlsi systems
The development of a high-level synthesis system for concurrent vlsi systems
SystemC Kernel Extensions For Heterogenous System Modeling: A Framework for Multi-MoC Modeling & Simulation
Hierarchical finite state machines with multiple concurrency models
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Hi-index | 0.00 |
Enhancing productivity for designing complex embedded systems requires system level design methodology and language support for capturing complex design in high level models. For an effective methodology, efficiency of simulation and a sound refinement based implementation path are also necessary. Although some of the recent system level design languages for system level abstractions, several essential ingredients are missing from these. We consider (i) explicit support for multiple models of computation (MoCs) or heterogeneity; (ii) the ability to build complex behaviors by hierarchically composing simpler behaviors; and (iii) hierarchical composition of behaviors that belong to distinct models of computation, as essential for successful SLDLs. These render an SLDL with modeling fidelity that exploits both heterogeneity and hierarchy and allows for simpler modeling and efficient simulation. One important requirement for such an SLDL should be that the simulation semantics be also compositional, and hence no flattening of hierarchically composed behaviors be needed for simulation. In this paper we show how we designed SystemC extensions to provide facilities for heterogeneous behavioral hierarchy, compositional simulation semantics, and implemented a simulation kernel which we show experimentally as up to 50% more efficient than standard SystemC simulation.