Incorporating cost modeling in embedded-system design

The emergence of multimedia and wireless applications as growth leaders has created an increased demand for embedded systems. Examples of such applications range from digital cellular telephones to high-performance avionics systems. Along with the increased market share of these products, however, has come an increase in system complexity and cost. For example, the complexity of embedded avionics systems has quickly grown from millions to billions of operations per second. These systems, formerly implemented in hardwired or uniprocessor architectures, now must consist of programmable-multiprocessor arrays to meet performance requirements. Time-to-market and life-cycle costs are key factors in the success of these products in the competitive electronics marketplace. These costs, therefore, should have a dominant influence on the design of embedded microelectronic systems. In addition, these systems must meet rigid form factor (such as size, power, and weight) constraints, which further complicate the design task. For designers of high-end embedded microsystems or large-volume consumer products, rapidly prototyping cost-effective implementations that meet stringent performance, functional, timing, and physical requirements is a formidable challenge     

The RASSP Program: Origin, Concepts, and Status: An Introduction to the Issue

The Rapid Prototyping of Application Specific Signal Processors (RASSP) program is a multi-year DARPA/Tri-Service initiative intended to dramatically improve the process by which complex digital systems, particularly embedded digital signal processors, are designed, manufactured, upgraded, and supported. This paper reviews the genesis of the RASSP program, considering both the problems that defined the need for the program, and the historical conditions under which it began. The RASSP program is then presented from two viewpoints. The first is programmatic, covering the goals and constraints of the program, and describing the roles of the various program participants. The second is technical, covering the major concepts upon which the developing RASSP approach to design is based and showing how the detailed technical discussions contained in the other papers in this issue of the Journal of VLSI Signal Processing relate to one another and fit into an overall development concept. The paper closes with a review of the status of the program as of this writing (Summer 1996).     

A case study in the development of multi-media educationalmaterial: the VHDL interactive tutorial

While industry and academia have been aware of the need for an intensive study of embedded digital system design, resource constraints, fuzzy objectives, and short-time horizon have handicapped progress. The $150M Rapid Prototyping of Application Specific Signal Processors (RASSP) program is a major DARPA/Tri-Service effort designed to overcome these limitations. This effort has developed a number of new technologies that will lead to dramatically shorter prototyping times and reduced life cycle costs. In a effort to ensure the successful transfer of these new technologies, the RASSP Education and Facilitation (RASSP E&F) program is working to transfer this technology into graduate and undergraduate curricula by developing and transferring educational material. Only by successfully inserting these rapid-prototyping technologies into the curricula and research activities of the university community will the long term benefits of these technologies be realized. To achieve this technology transfer objective, the RASSP E&F program has developed educational material on the key elements of rapid-prototyping of embedded digital systems technology. The result of this effort is a set of educational “modules” covering selected topics. In conjunction with IEEE, several of the modules developed by the RASSP E&F program designed to teach students the VHDL language were converted to HTML and linked with the VHDL language reference manual to form a self-paced VHDL interactive tutorial. A demonstration version of this learning tool, currently published by IEEE, is presented     

A dynamic approach to controlling high-level synthesis CAD tools

In this paper, we present a novel model of CAD tool control that can be used in the constraint-directed control of high-level synthesis tools. To enable this control we introduce the concept of a design space reasoning mechanism. We formally describe a statistical based, machine learning process that automatically generates the tool control knowledge necessary to drive the design space reasoning mechanism. The representation of this tool control knowledge in the form of a fuzzy, linear differential, qualitative model is described. Finally, the experimental results obtained using the Magellan system are presented