ECE 562

Digital communication systems modulation, demodulation, signal space methods, channel models, bit error rate, spectral occupancy, synchronization, equalization, trellis-coded modulation, wireless channels, multiantenna systems, spread spectrum, and orthogonal frequency modulation. Course Information: 4 graduate hours. No professional credit. Prerequisite: ECE 461 or ECE 459.

Communication Systems are the basic workhorses behind the information age. Examples include high speed communication networks, wireless and wireline telephone systems, high speed modems, etc. The basic currency of information is digital: bits . Broadly speaking, this course is centered around a single theme: reliably communicate bits over an unreliable physical medium. The emphasis is on how to transfer this currency between a transmitter-receiver pair. The transfer involves a physical medium, whose input-output characteristics are not deterministically known. The curriculum has three broad parts:

• Channel Model : Even though the input-output relationship of the physical medium is not deterministically known, statistical quantities of this relationship, such as mean and correlation, can be physically measured and are typically constant over the time-scale of communication.
• Transmission and Reception Strategies : The statistical model of the physical medium is then brought into bearing in the engineering design of appropriate transmission and reception strategies (modulation and demodulation, in the language of this course).
• Design Resources and Performance : The resources available to the communication engineer are {\em power} and {\em bandwidth}. The final part of the course is to relate the statistical performance of the communication strategies to these two key resources.

These three parts are discussed in the course in the context of four important channel models:

• Erasure channel: The packet sent is either received correctly or is erased (known to be incorrectly received). This is a basic model that provides a first order view of the more complicated wireline channel.
• Additive white Gaussian noise channel: The received signal is the transmit signal plus a statistically independent signal. This is a basic model that underlies the more complicated wireline and wireless channels.
• Telephone channel: The received signal is the transmit signal passed through a time-invariant, causal filter plus statistically independent noise. Voiceband v.90 modem and DSL technologies are used as examples.
• Wireless channel: The received signal is the transmit signal passed through a time-varying filter plus statistically independent noise. The GSM and 1xEV-DO standards are used as examples.

Undergraduate digital communication and probability courses and some discrete-time signal processing background.

2/13/2013