Interoperability of Special-Purpose Tools with Ptolemy

This goal of this project is to connect special-purpose symbolic analysis tools (e.g. Mathematica) to aid in computing parameters and rearranging systems, and special-purpose numeric processing tools (e.g. MATLAB) to aid in simulation. MATLAB is commonly used for prototyping signal processing algorithms because of its signal and image processing toolboxes and its visualization capabilities. MATLAB algorithms can be imported into Ptolemy. In Ptolemy, a MATLAB block represents a sequence of MATLAB commands to be applied to the block's input signals to produce output signals.

Mathematica is a symbolic algebra system. By itself, it can assist designers in deriving and analyzing values of free parameters, especially when the values of parameters are defined in terms of other parameters that are not known until run time. By using the signal processing packages [3][4] for Mathematica, designers can apply rules to generate new and redesign existing synchronous dataflow graphs. (Synchronous dataflow is a computational model which processes computations on uniformly sampled data.) We are working on exchanging system descriptions between Ptolemy (which uses flow graphs) and the Mathematica signal processing packages (which uses algebraic equations).

1. A. Kalavade, J. L. Pino, and E. A. Lee, "Managing Complexity in Heterogeneous Specification, Simulation, and Synthesis," invited paper, Proc. of IEEE Int. Conf. on Acoustics, Speech, and Signal Processing, May, 1995, Detroit, MI.
2. B. L. Evans, A. Kamas, and E. A. Lee, "Design and Simulation of Heterogeneous Systems Using Ptolemy," First Annual Rapid Prototyping of Application Specific Signal Processors (RASSP) Conference, Arlington, VA, Aug. 15-18, 1994, pp. 97-105.
3. B. L. Evans, J. Teich, and C. Schwarz, "Automated Design of Two-Dimensional Rational Decimation Systems," Proc. of IEEE Asilomar Conf. on Signals, Systems, and Computers, Oct. 31 - Nov. 2, Pacific Grove, CA, 1994.
4. B. L. Evans, "A Knowledge-Based Environment for the Design and Analysis of Multidimensional Multirate Signal Processing Algorithms", Georgia Institute of Technology, Atlanta, GA, June, 1993.