CSE4210 Architecture and Hardware for Digital Signal Processing

Description

The field of DSP (digital signal processing)is driven by two major forces, advances in DSP algorithms, and advances in VLSI technology that implements those algorithms. This course addresses the methodologies used to design custom or semi-custom VLSI circuits for DSP applications. Several techniques, such as pipelining, retiming, parallel processing, and folding, are introduced for speed enhancement, area reduction, and power saving.

Learning Outcomes

Upon completion of the course, students should be able to:

  1. map a DSP algorithm to a graphical representation and determine its fundamental lower bound on the achievable iteration or sampling period;
  2. use pipelining, retiming, and parallel processing to improve the performance of a DSP implementation;
  3. use folding technique to reduce silicon area in a DSP implementation;
  4. assess alternative architectures based on a given set of design specifications;
  5. implement a DSP algorithm based on an optimized architecture.

Instructor & Office Hours

Lecture and Lab Times

Teaching Assistant