Lecture 1 Notes
Lecture 1 Notes
Is it required? Yes. You will use this expensive text for two courses, a fact that
should help with the cost.
Marking Scheme:
DCN230 is for all intents and purposes an introductory data communications course
disguised under the course title “Protocols: Standards and Models”. This title
reflects the fact that virtually all data communications courses introduce concepts
in relation to the two best known models called the OSI and the TCP/IP or Internet
Model as it is often called.
Why do we have two models? The OSI model was developed in 1970s by the ISO as
the first systematic attempt to describe networked communications. An open
system is a set of protocols that allows any two different systems to communicate
regardless of their underlying architecture (Forouzan). We will define a protocol as a
rule, or set of rules that govern communication. For example, the protocol that
governs communication in this classroom involves the use of English, a certain
volume or voice level, the custom of raising one’s hand to ask or answer a question,
etc. The OSI Model is technically not a protocol, then. It is a model for
understanding and designing a network architecture that is flexible, robust, and
interoperable (Forouzan).
When the internet was launched it quickly became apparent that the protocol
actually in use was that of the TCP/IP model, which had existed prior to the OSI
Model. The TCP/IP protocol suite is the dominant protocol family that governs
internet communications. We might say, then, that the OSI model is a theoretical
reference model, while the TCP/IP or Internet Model is the practical implementation
of the layers described in the OSI model. We will examine the OSI layers today and
compare them with those of the Internet Model, but after today, all references will
be to the Internet Model.
Note that I will present the Internet Model as a four-layer model but that some
accounts involve a five-layer model. Unfortunately, your textbook presents the five
layered version. The difference is that Fourouzan retains the use of the OSI layers
one and two, namely, the physical and data link layers, while my (more popular
account) combines the function of these two layers and calls this entity the
“Network Access Layer”. My account is also consistent with the Cisco curriculum
that some of you may wish to pursue certification in, so my account may prove to
be more useful.
The benefits of using a layered model include
• Fosters competition
The following figure demonstrates the parallels between the OSI and the Internet
Model layers:
Note the numbering of the layers from highest to lowest – i.e. the “highest” layer or
the topmost layer has the highest number. Note the number and the ordering of
the layers, because as the semester progresses, we will often refer to layers by
number rather than name.
Layer Functions
Note that the following list names, but does describe details of the function of each
of the OSI layers. See your text for detailed descriptions.
Framing
Physical addressing (MAC)
Flow control (e.g. fall forward, fall back)
Error control (CRC)
Access control (MAC)
Logical addressing
Routing
Dialog control
Synchronization
Brief summary of TCP/IP Model Layer Functions (Note: inadequate for test
purposes)
Encapsulation and Protocol Data Units (PDUs)
The term encapsulation refers to the transformation of data as it makes its way
down through the OSI layers. Headers and trailers are added as the data moves
from one layer to the next. These are stripped off at the receiving end as the data
moves up through the layers until it is presented to the user in its original format at
the application layer. We will be using the network simulation software called
Packet Tracer this semester. This software enables us to examine PDUs and their
contents in a simulated network environment and should help to bring life to these
concepts.