Modem Design for Today’s Digital Communications

A 3-Day Short Course

Wireless modem design has changed substantially with emerging technologies yielding power and spectral efficiencies unknown only a few years earlier. This course equips wireless modem and receiver designers with the latest information on equalizers, digital modulation, acquisition techniques, and forward error coding design. Fundamental concepts and design considerations for modem design are blended with the latest technologies and concepts, such as MIMO, LDPC codes, and synchronization techniques.

Complete Details

This course is intended for scientists, systems engineers, and software/hardware engineers who are designing or analyzing wireless modems and need to understand the trade-offs and advantages of these new architectures. Design philosophies consistent with Software Defined Radio (SDR) requirements are discussed, along with operation within Time Division Multiple Access (TDMA) and Code Division Multiple Access (CDMA), and Mobile ad Hoc (MANET) networks.

The course thoroughly examines power and spectral efficient modulation/coding to allow maximum exploitation of an increasingly crowded frequency spectrum. Modeling of the physical channel is discussed and techniques for modeling multipath and scintillation are presented. Participants can then optimize the physical layer of their modem with the latest waveforms in digital modulation, such as Continuous Phase Modulation (CPM), Orthogonal Frequency Domain Modulation (OFDM), spread spectrum, 8+4 PSK, etc.

The course establishes a design methodology based upon the fundamentals of estimation theory applied to symbol detection and synchronization instead of ad hoc algorithmic design. Traditional detection and synchronization techniques for modulations, such as BPSK/QPSK, FSK, and QAM, are presented in addition to the latest techniques in maximum likelihood detection, equalizers, and channel state estimators. Because modems are often required to support legacy waveforms, such as Amplitude Modulation (AM) or digital Frequency Modulation (FM), techniques are presented for these waveforms as well.

The instructor’s specialty is acquisition and synchronization and establishes an underlying theme of the course: acquisition of signal is fundamental. A novel FFT acquisition technique is introduced where frequency, phase, and timing variables are simultaneously computed from a single FFT. Other synchronization methods and techniques for single-carrier modulations, including spread spectrum, are presented and analyzed. Acquisition of OFDM and SC-FDM waveforms is discussed.

Modem architectures, such as direct conversion, IF-sampling, and super heterodyne, are discussed and analyzed. To support wideband software radios, digital resampling is presented with digital interpolation techniques. Several methods of equalization are studied, covering simple Least Mean Square (LMS), Recursive Least Squares (RLS), constant modulus, and Decision Feedback Equalizers (DFE). Forward error coding techniques, such as convolutional, Reed Solomon, Turbo Codes, Low Density Parity Codes (LDPC), and others are presented.

Real-time MATLAB and Mathematica simulations/analysis embedded within the lectures provide new insights on modem techniques ranging from digital interpolation to soft output Viterbi algorithms. Participants practice hands-on simulations of wireless modem algorithms in a computer laboratory.The course should enable participants to:

  • Design a modem architecture, such as superheterodyne, direct conversion, or IF-sampling
  • Understand TDMA, CDMA, and MANET wireless networks
  • Utilize delta-sigma algorithms and circuits
  • Compute error bounds synchronization estimators
  • Generate an Eb /N0 degradation budget for all of the modem functions
  • Design lock detectors for modem synchronization
  • Design carrier phase and symbol tracking loops
  • Design feed-forward synchronization implementations
  • Design timing interpolators (re-samplers)
  • Design Viterbi-based detectors and Soft Output Viterbi Algorithms (SOVA)
  • Combine modulation and forward error coding for improved spectral efficiency
  • Specify interleavers for concatenated forward error coding
  • Specify preambles for modem acquisition
  • Use FFT acquisition techniques
  • Design equalizers and channel state equalizers for fading channels
  • Design digital implementations of analog modulators and demodulators
  • Specify and design automatic gain control (AGC) loops
  • Design spread-spectrum demodulators

It is recommended that participants take the companion course, Phase-Locked Loops for Digital Communications, for a thorough understanding of these technologies, although this is not a prerequisite.

Course Materials

Lecture notes are distributed on the first day of the course. These notes are for participants only and are not for sale. A disk of laboratory exercises in MATLAB also is provided.

Coordinator and Lecturer

Donald R. Stephens, PhD, Chief Scientist, CommLargo, Inc., St. Petersburg, Florida. Dr. Stephens is a member of the JPEO JTRS technical staff. In addition to JTRS activities, he develops spectrally efficient waveforms for MILSATCOM and other military communication systems. To implement the waveform designs, Dr. Stephens also develops novel modem architectures and algorithms exploiting communications theory and signal processing. With companies such as Raytheon, E-Systems, McDonnell Douglas, Emerson Electric, and Scientific Atlanta, he has developed multiple communications and radar receivers. These systems have included CPM, spread spectrum waveforms, wavelet video compression, and multi-spectral signal processing. Dr. Stephens also has participated in a joint government/industry MILSATCOM working group on Demand Assigned Multiple Access (DAMA). He is the author of the text, Phase-Locked Loops for Wireless Communications: Digital, Analog, and Optical Implementations, Second Edition (Kluwer Academic Publishing, 2002), and previously taught electromagnetic theory and digital signal processing as an adjunct professor at Southern Illinois University, Edwardsville.

Daily Schedule

Day 1

Introduction of the course participants and a brief description of the class topics. Introduction to modem architectures, such as low-IF, direct conversion, IF-sampling, and superheterodyne configurations. Software programmable radio architectures and strategies. TDMA, CDMA, and MANET wireless networks. Modem link layer topics, such as interleaving and forward error coding.Introduction to the fading channel and mitigation strategies/algorithms. Equalizer design including LMS, RLS, and Viterbi implementations.

Eb /N0 budget and specification of other modem parameters. Introduction to digital modulation waveforms, such as BPSK, QPSK, CPM, OFDM, etc. Combined modulation and coding, such as Ungerboek’s trellis codes and implementation. Introduction to spread spectrum: frequency hopping and direct sequence. Review of legacy analog waveforms and modem implementation algorithms.

Day 2

Introduction to estimation theory. Introduction to the Receiver Operating Curve. Review of Bayesian, ML, and MAP estimation techniques. Development of the Cramer-Rao lower bound and application to modems. Carrier and symbol tracking algorithms for digital modems. Design of synchronization lock detectors. Introduction to the Hilbert Transform and implementation in all-digital modems. Analysis and demonstration of timing interpolators.

Advanced digital designs. Maximum Likelihood (ML) techniques for continuous phase modulation (CPM) modems. GSM modems. Joint carrier/phase estimation algorithms. Frequency Shift Keying (FSK) demodulators. Forward error encoding techniques, including convolutional, Reed-Solomon codes. Introduction to MIMO, including maximal ratio combining, Alamouti, and space-time coding.

Hands-on simulation of modems and algorithms using MATLAB in the computer laboratory.

Day 3

Viterbi demodulators and Soft Output Viterbi Algorithm(s) (SOVA) as applied to turbo and LDPC codes. Spread spectrum receivers and synchronization techniques. Introduction to wireless networking. Design of channel state estimators and equalizers. Sub-fractional sampling (interpolators) algorithms. Automatic gain control (AGC) loop design and analysis. Phase noise effects on modem performance. Introduction to the LTE waveforms. Synchronization techniques for OFDM modems.

Hands-on simulation of modems and algorithms using MATLAB in the computer laboratory.

For more information contact the Short Course Program Office:
shortcourses@uclaextension.edu | (310) 825-3344 | fax (310) 206-2815