Data Compression and Modeling

(3-0-0-3)

CMPE Degree: This course is Not Applicable for the CMPE degree.

EE Degree: This course is Not Applicable for the EE degree.

Lab Hours: 0 supervised lab hours and 0 unsupervised lab hours.

Technical Interest Group(s) / Course Type(s): Digital Signal Processing

Course Coordinator:

Prerequisites: ECE 4270

Catalog Description

Theory and algorithms of signal encoding and decoding for data compression. Applications in information systems, digital telephony, digital television, and multimedia Internet.

Textbook(s)

Introduction to Data Compression, Vector Quantization and Signal Compression, Digital Compression for Multimedia, Principles and Standards

Course Outcomes

Not Applicable

Strategic Performance Indicators (SPIs)

Outcome 1 (Students will demonstrate expertise in a subfield of study chosen from the fields of electrical engineering or computer engineering):
1. Understand the theory and practice of data compression and modeling as well as its applications in digital media and communication systems in everyday use.

Outcome 2 (Students will demonstrate the ability to identify and formulate advanced problems and apply knowledge of mathematics and science to solve those problems):
1. Understand and implement methods and algorithms for data compression and modeling

Outcome 3 (Students will demonstrate the ability to utilize current knowledge, technology, or techniques within their chosen subfield):
1. Design algorithms to model a given signal, extract model parameters, and achieve effective data compression on the signal

Topical Outline

Introduction
signal compression, lossless and lossy compression
communication systems and building blocks: sources, channels, and codes
issues - fixed rate and variable rate, robustness to channel errors, degradation and perceptual effects

Quantization theory
uniform quantization, distortion and bit rates
amplitude distribution and high-rate quantization theory
Bennett approximations and optimal performance, Lloyd's code optimality and algorithm
elementary distortion-rate theory

Architecture for data compression & introduction to data modeling
signal models & spectral analysis
quantization with memory
fixed-rate vs. variable-rate code
entropy, estimated entropy, complexity and typical sequence of an ergodic source
variable rate quantization: lossless codes, prefix code

Lossless Coding Techniques
Huffman coding, arithmetic coding
Universal lossless codes, adaptive and predictive lossless coding

Distortion & Similarity Measures
sample difference, sum of squared deviations and Euclidean distance
Lp-norm, city-block distance, Mahalanobis distance
transformation and transformation invariant similarity measures
spectral distortion measures
mutual-information, divergence, and Kullback-Liebler number
perceptual issues

Coding algorithms scalar quantization
clustering algorithms for quantizer design
the Lloyd algorithm and its generalization
entropy-constrained quantizers

Coding algorithms - vector quantization (VQ)
sphere packing and optimal uniform lattice quantizers
Lloyd algorithm - revisited
progressive vector quantization
variations of vector quantization
finite-state VQ and Markov models
tree and trellis encoding

Applications
speech and audio coding
image and video coding

Compression standards and formats
Historical and evolutional aspects behind development of standards
Application areas