Department of Electrical and Computer Engineering MS Thesis Defense

SHARC WDF Library and R-Solver – Tools for Prototyping Wave Digital Filter Models of Circuits

Samuel Schachter

Supervised by Professor Mark Bocko

Thursday, June 24, 2021
10:30 a.m.

Join Zoom Meeting
https://rochester.zoom.us/j/95315817356

Abstract:

Wave Digital Filters (WDF) represent a burgeoning approach to digitally modeling analog circuitry. WDFs simulate the wave scattering relationship between discrete circuit elements, allowing for a computationally efficient simulation of a circuit’s behavior in real-time [1]. In recent years, researchers have made great strides in expanding the class of circuits that can be modeled using WDFs. One such development is the introduction of R-type adaptors, which are used to represent circuit topologies that cannot be broken down into simple serial or parallel connections [2].

In this thesis, I present two tools to aid in the prototyping and derivation of WDF models. The first is a C library of Wave Digital Filter objects which is optimized for use with the SHARC Audio Module, a floating-point DSP processor designed for audio software prototyping [3]. The second is R-solver, a MATLAB live script that algorithmically derives the scattering parameters for any R-type adaptor [1].

Chapter 1 presents the background information necessary to form WDF structures and implement them as a real-time traversable connection tree. Chapter 2 discusses the SHARC WDF library and the specific challenges of creating Wave Digital Filter models for the SHARC. Finally, Chapter 3 presents R-solver and how it may be used to simplify the WDF model derivation process.