This course addresses the design and evaluation of digital communication systems, emphasizing areas of new technology, such as turbo codes and low-density parity-check (LDPC) codes.
- Chip designers
- Engineering managers involved in the design, planning, implementation, or testing of communication systems
The starting point and frame of reference in this course is quite different than in a basic course that outlines fundamental relationships and shows how to compute performance parameters of given systems. This course begins only with some requirements—the focus is not on a particular system but on how to make reasonable design choices based on given requirements. These requirements then drive us toward some candidate systems.
Chief Goals of the Course
- Learn to make good design choices in tough situations. For example, if we are short of bandwidth, power, or both, what do we do?
- Learn about channels—the Good, the Bad, and the Awful. Multipath and fading are described in a very organized way, as well as information on learning to live with their degrading effects.
- See how the predictions of Shannon have been met in the last dozen years through iterative decoding—turbo codes and LDPC codes.
- Play the “LDPC game” for teaching the basic ideas behind the message-passing algorithms used in turbo and LDPC codes. Participants represent either parity-checks or data-bits, then pass probability messages to each other, thereby gaining insight as they perform a step-by-step enactment of these important concepts.
- Examine other creative techniques, such as trellis-coded modulation and Orthogonal FDM (OFDM)
- Present it all in an easy-to-understand way
Please bring a calculator.
Lecture notes are distributed on the first day of the course. These notes are for participants only and are not for sale.
Coordinator and Lecturer
Bernard Sklar, PhD, President, Communications Engineering Services, Tarzana, California. Dr. Sklar previously was at The Aerospace Corporation and has acquired over 50 years’ experience in the aerospace/defense industry in a wide variety of technical design and management positions. He has worked at Republic Aviation Corporation, Hughes Aircraft Company, and Litton Systems, and has taught communications at both the University of Southern California and UCLA. He also has taught at other universities, and has presented numerous short courses throughout the United States and Europe. Dr. Sklar has published and presented scores of technical papers, is the recipient of the 1984 Prize Paper Award from the IEEE Communications Society for his tutorial series on digital communications, and is the author of Digital Communications: Fundamentals and Applications, Second Edition (Prentice-Hall, 2001). He is a past chairman of the Los Angeles Council IEEE Education Committee.
Defining, Designing, and Evaluating Systems
A systematic approach for meeting bandwidth- and error-performance requirements; criteria for choosing modulation and coding schemes for various channel types. Given only the system requirements, how does one select appropriate signaling methods? The subtle computations required and the step-by-step approach used for evaluating most any digital communication system. Software examples and in-class training exercises.
How I Learned to Love the Trellis
Finite state machines, partial response signaling; how the Viterbi algorithm can uniquely operate as an equalizer, a detector, or a decoder; likelihood functions; bit-by-bit detection versus sequence estimation; detecting and decoding signals with memory; hard versus soft decisions, add-compare-select (ACS) architecture; similarities between convolutional decoding and equalization; applying the Viterbi equalizer in GSM.
Turbo Codes and the Map Algorithm
Concatenated codes and the fundamentals of turbo codes; extrinsic information; iterative decoding and near-Shannon limit performance; the MAP algorithm, how it differs from the Viterbi algorithm, and how it is implemented to yield bit-by-bit a posteriori probabilities; numerical examples of turbo decoding.
Traditional coding methods are disappointing when applied to bandwidth-limited channels. How do trellis-coded modulation (TCM) schemes differ from error-correction codes and how do they achieve coding gain without expending bandwidth? Objectives of TCM; code-to-signal mapping; signal set-partitioning; trading complexity for improved error-performance; TCM for various signal types; multidimensional-trellis coding. In-class training exercises.
Low-Density Parity-Check (LDPC) Codes
Fundamentals of LDPC codes; Tanner graphs; bit nodes, check nodes, and the message-passing algorithm; iterative decoding and how these codes asymptotically approach the Shannon limit. Comparison with turbo codes and software exercises. The “LDPC game” is played, which facilitates a rapid understanding of the basic ideas behind the message-passing algorithm.
Fading Channels: Characteristics and Mitigation
Large- and small-scale fading, and their mechanisms. What are the most serious degradation effects of fading? Signal dispersion, channel-induced intersymbol interference (ISI). What is the difference between frequency-selective fading and flat fading? Fast fading and slow fading? Characterizing the channel as a time-varying filter; channel coherence bandwidth; time-variant structure of the channel due to motion; channel coherence time; Doppler spreading. In a fading channel, why is signal dispersion independent of fading rapidity? Performance degradation effects: loss in SNR, irreducible bit-error-rate, ISI distortion, pulse mutilation, Doppler spreading. How to design a system that can withstand fading. Different effects call for different mitigation types. Why does orthogonal FDM (OFDM) appear promising for wideband wireless systems? Training exercises.
CDMA Mobile Telephony
FM versus TDMA versus CDMA. How did CDMA (a relative newcomer) win its way into third-generation standards? Interference suppression in CDMA; cellular structure; IS-95 requirements: forward channels (pilot, synch, paging, and traffic); open- and closed-loop power control; reverse channels (access and traffic); spreading codes, orthogonalizing codes; diversity types: time, path, and spatial; using the Rake receiver to exploit the effects of multipath.
Mobile Telephony: WCDMA for UMTS/IMT-2000
Improved capacity and coverage for third-generation systems, comparison with second-generation systems; maximal length codes, composite codes: Gold codes, variable length orthogonal codes and variable rate transmission, multiple spreading and scrambling; inter-cell synchronous and asynchronous operation, fast cell search, fast power control, interference cancellation, adaptive antenna array techniques.
For more information contact the Short Course Program Office:
firstname.lastname@example.org | (310) 825-3344 | fax (310) 206-2815