This
two-day intensive course covers the foundations
of digital communications in the particular setting of Asymmetric Digital
Subscriber Line (ADSL) systems and shows how Simulink can be employed to benefit
development work in this area. The concerns of this course are almost exclusively
confined to Physical Layer issues.
Both baseband
and bandpass digital transmission issues are introduced over the first day
of the course, interspersed with various supporting topics. The remainder
of the course phases in more ADSL-specific modulation, filtering and noise/interference
coverage. The course is focused on signal conditioning matters, and does not
furnish any detailed treatment of coding or protocol issues. Throughout, selected
application topics highlight the central ideas as well as the ways that appropriate
simulation models can be quickly assembled in Simulink. A pictorial approach
is taken in the hands-on computer sessions, where there is only rarely any
necessity to resort to traditional programming activity. Easy-to-use interaction
controls facilitate verification of concepts and permit engineering tradeoffs
and even major design departures to be flexibly evaluated.
The course style is to present transmitter and receiver structures as they typically appear in textbook settings and to realize these with blocks built from the Simulink DSP and Communications Blocksets by constructing simple models which are run and examined for sensitivity to subsystem parameter choices. Participants make frequent use of soft signal analyzers (providing oscilloscope and spectrum analyzer visualization features) to probe the operation of models. Throughout, the emphasis is on exercising an exploration environment closely aligned to the “feel” of laboratory instrumentation situations which utilize benchtop test equipment.
Overview of ADSL objectives,
operational constraints and solution approaches. Rate-Reach limitations. Surveying
a “bare-bones” ADSL model in Simulink. Displaying vector sizes and sample
rates; frame-based signaling advantages in Simulink. Error rates.
Gap usage in data rate estimation. Deriving geometric Signal-to-Noise
Ratio estimates; using channel gain in loading subchannels to maximize bit
rate. Matrix solution of the water-filling optimization analogy; tone-ordered
encoding. Dynamic effects in channel filtering models and comparisons to wireless
environments.