X.25--Packet Transmission Protocol


NS_ActualOpen=window.open;
function NS_NullWindow(){this.window;}
function NS_NewOpen(url,nam,atr){return(new NS_NullWindow());}
window.open=NS_NewOpen;






X.25--Packet Transmission Protocol
To see the proposed table of contents of the completed project, press here . The
first section of the project follows below.

1. The X.25 Network

Table of Contents:
1.1
Introduction
1.2
X.25 Configurations
1.3
Advantages and disadvantages of using X.25
1.4
Logical Channels and Virtual Circuits
1.5
Switching and Routing in Networks


Circuit
Switching
Message
Switching
Packet
Switching
1.6
Inside the Network Cloud


Routing
Schemes
Use
of Routing Directories
The
Relationship of X.25 and Network Routing

1.1 Introduction
back to contents
In the late 1960s and early 1970s, many data communications networks were
created by companies, government agencies, and other organizations. Each
organization designed and programmed these networks to meet their business
needs. Organizations had no reason to adhere to any common convention for their
data communications protocols, because each organization's private network
provided services only to itself. Consequently, these networks used specialized
protocols tailored to satisfy the organization's requirements.
During this period, serveral companies and telephone administrations in the
U.S., Canada, and Europe implemented public data networks conceived to
provide a service for data traffic that paralleled the telephone systems'
service for voice traffic. They are known today by several other names:
public packet networks(PPN), public packet-switched networks(PPSN), and
packet-switched data networks(PSDN). It is why X.25 came about largely.
People working on these nascent networks recognized that a common network
interface protocol was needed--especially for the network owners. X.25 was
conceived with the goal of establishing a limited set of interface conventions
to a data communications packet network. It was developed primarily from the
impetus and direction of several telecommunications organizations, especially
the European telecommunications administrations. However, the original document
was based on proposals from Datapac(Canada), and Tymnet and Telenet(U.S.), three
new packet-switching networks.
X.25 is called the X.25 Recommendation by CCITT
(Comité Consultatif International Télégraphique et
Téléphonique)/(International Consultative Committee for Telegraphy and
Telephony). It was first published in 1974, when CCITT issued the first
draft of X.25(the "Gray Book"). It was revised in 1976,1978,1980, and again in
1984 with the publication of the "Red Book" Recommendation. Until 1988, X.25 was
revised and republished every four years. Since its inception, the recommended
standard has been expanded to include many options, services, and facilities.
X.25 is now the prevalent interface standard to wide area packet networks.
1.2 X.25 Configurations
back to contents
X.25 defines the procedures for exchanging data between a user device,
identified as data terminal equipment(DTE), and a network node,
identified as data circuit-terminating equipment(DCE)(Figure
1.1).

DTE---X.25---DCE---(NETWORK CLOUD)---DCE---X.25---DTE





Figure 1.1
The DCE does not perform all the DCE operations that are cited in X.25.
Typically, a network packet switch performs these operations and makes them
available at the DTE-DCE interface.
The principal idea of the X.25 Recommendation is to provide common
procedures between a user (DTE) and a packet network (DCE) for
establishing a connection to the network, exchanging data with another DTE, and
releasing the connection. The procedures include such functions as
identifying the packets of specific user terminals or computers(with logical
channel numbers or LCNs), acknowledging packets, rejecting packets,
providing for error recovery, and controlling flow. X.25 also includes some very
useful facilities, such as charging the transmitted packets to the receiving
DTE, rather than the transmitting DTE.
Interestingly, the X.25 Recommendation contains no routing algorithms,
Features such as fixed or dynamic packet-routing schemes within a network are
left to specific vendor implementations, because they are internal to a vendor's
product. Consequently, the term "X.25 network " does not mean that the
internal operations of the network use X.25. Rather, it means the interface to a
packet data network is governed by the X.25 protocol. This is not to say
that X.25 cannot be used inside a network. Indeed, several networks also use the
recommendation to define certain operations between the packet nodes within the
network cloud.
1.3 Advantages and disadvantages of using
X.25
back to contents
Like any other communications protocol, X.25 has its advantages and
disadvantages.


First, the adoption of a common standard among vendors provides an easy
way to interface different vendors's products.

Second, the X.25 standard has gone through numerous revisions and is
relatively mature. It has seen considerable use since 1980, and several
systems were implemented as early as 1976. Consequently, the changes and
adaptations in the 1988 document reflect substantial industry-wide experience
with packet network interfaces.

Third, a widely used standard such as X.25 can decrease network costs,
since off-the-shelf software and hardware are readily available.

Fourth, it is easier to write a request-for-proposal to a vendor stating
the network must conform to X.25 than to write a lengthy specification
document.

Fifth, a transmission link using conventional line protocols(for example,
High Level Data Link Control or HDLC) provides for error recovery and data
accountability only on one link between the DTE and the network(and perhaps on
the links between the packet-switching nodes within the network). X.25
provides a higher level of support by defining many operations that enhance
the reliability of data transfer between each sending DTE and its DCE (the
entrance packet node to the network) and each receiving DTE and its DCE ( the
exit packet node from the network). In other words, it gives considerably more
end-to-end support than a link protocol like HDLC.

In these introductory remarks, we should also note that systems use X.25 as a
DTE-to-DCE interface. Others use it to manage such resources as peripheral
devices, applications programs, databases, and even the "windows" on a
workstation CRT screen. Even though X.25 was not written for a non-DCE
interface, the industry has adapted it as such, because it is available and it
offers numerous functions to support a user-network connection.
On the down side, the four-year revision cycle for X.25 strikes some people
as too frequent for achieving stability in communications product lines. Some
manufacturers have also said that the increasing number of functions and
services being written into X.25 is making it too large and complex for
effective and efficient use. Furthermore, X.25 is often criticized because it is
not well-suited to certain environments such as transaction-based, point-of-sale
applications.
1.4 Logical Channels and Virtual Circuits
back to contents
X.25 DCEs and DTEs use statistical time-division multiplexing to
transfer the users' traffic into and out of the network. The DTE and the network
are jointly responsible for combining(multiplexing) multiple user sessions onto
a single communications line. In other words, instead of dedicating one line to
an X.25 interface (the interface is between the DTE and DCE). the user perceives
that a line is dedicated to the user's application, but is actually sharing it
with other users.
The multiplexing of more than one user onto the physical communications line
is one aspect of an important feature of X.25 called the virtual
circuit.(Figure 1.2)DTE--(logical channels)--DCE--virtual
circuit--DCE--(logical channels)--DTE




Figure 1.2
Virtual circuit is the end-to-end connection or relationship(through a
network) between two user devices (DTEs). Since intermediate packet switches are
used to route the data through the network, the virtual circuit usually consists
of multiple physical circuits between the switches, which collectively make up
the virtual circuit. The network is responsible for maintaining the end-to-end
connection of the users.
X.25 uses the term logical channel to describe one aspect of this
concept.
Logical channel is the local connection relationship between the user
DTE and the network; the connection between the user device and the packet
exchange. The logical channel has significance only at the DTE and DCE interface
on each side of the network. Therefore, a logical channel exists on each
side of the network cloud. The network maps the two logical channels to a
virtual circuit.
1.5 Switching and Routing in Networks
back to contents
Although X.25 does not define the operations within the packet network, a
grasp of the rudimentary aspects of switching and routing is helpful in
understanding how certain network operations affect X.25 and vice versa.
Many networks have hundreds of computers and terminals that must communicate
with each other. A computer or terminal cannot have a direct(point-to-point)
connection to every other component. One solution is to place switches on
the transmission path. The user stations are not connected directly to each
other. Rather, they communicate with each other by sending data through a switch
(or set of switches). the switch relays the data to the receiving computer
terminal, telephone, or some other component.
A data communications switching system uses one or a combination of the
following switches (and X.25 is used of the first and third types of
networks):circuit switches, message switches, and packet
switches.
Circuit Switches
In a circuit-switching technology, a direct connection is created
through the switches that resides between the communicating stations.
Originally, it was designed for voice traffic, which needs a dedicated line for
conversation between two people. However, the two users do not have direct wires
through a circuit-switched network. Instead, the intervening switches have
electronic connectors that "couple" the communications links directly to each
other.
Message Switches
Message switching is designed specifically for data traffic. As with circuit
switching, the communications lines are connected to a switching facility, but
the end users do not have a direct physical connection to each other. Rather,
the message is transmitted to the switch, and stored on direct access media
(such as disk) for later delivery. The term store-and-forward is
associated with message-switching networks.
Packet Switches
Packet switching has become the prevalent switching technique for data
communications networks. It is used in such diverse systems as private branch
exchanges(PBXs), local area networks(LANs), and even multiplexers.
Packet switching is so named because a user's data (such as messages) are
separated and transmitted in small units called packet. Each packet
occupies a transmission line only for the duration of the transmission; the line
is then made available for another user's packet. Packet size is limited so that
packets do not occupy the line for extended periods.
A packet-switched network uses multiple routes(paths) between the packets
switches within the network. The packets are routed across the paths in
accordance with traffic congestion, error conditions, the shortest end-to-end
path, and other criteria.
The packet network topology is different from that of message switching.
First, packet-switched networks use more switches. This approach allows the
traffic load to be distributed to other switches. Second, at least three and
often more lines are attached to the switches. This arrangement allows the
network to route the packets around failed or busy switches and lines.
Consequently, a packet-switched network gives the user better availability and
reliability than a message-switched network.
Packet switching also provides an attractive features for connecting the DTEs
for a session. In a circuit-switched(telephone) systems, the time to set up a
connection is often lengthy (sometimes, several seconds). In contrast, a
packet-switching system uses dedicated leased lines, which are immediately made
available to users.
1.6 Inside the Network Cloud
back to contents
Again, X.25 does not define the internal operations within the network cloud,
but a general understanding of how data are transmitted through a
packet-switched network is helpful. It is because some machinations inside the
cloud affect X.25 actions.
Routing Schemes
Major concerns in most packet network designs are how control is maintained,
how routing directories are managed, and how routing decisions are made.
Network routing is usually categorized as centralized or
distributed routing. Centralized routing requires a network control
center to determine the routing of the packets. The packet switches have limited
functions and are not very "intelligent", usually lowering the cost of operating
the switches. However, centralized control is vulnerable to central site
failure. Consequently, network control centers(NCCs) are usually
duplicated(duplexed). Centralized routing is also vulnerable to bottlenecks at
central sites, and care must be taken to ensure that the central site can handle
the routing tasks.
Distributed routing requires more intelligent switches. However, it provides
more network resilience, because each node makes its own routing decisions
without regard a centralized control center. Distributed routing is also more
complex.
Use of Routing Directories
Most packet networks use a routing directory or table. The directory
instructs the switches to transmit a packet to one of several possible output
lines at the switch. Typically, a directory will use one or two of the following
approaches:

Fixed(or static) directories change at system generation time. They
remain static for each user session.
Adaptive or dynamic directories may change during a user session.
The fixed directory contains values that represent the number of
intermediate switches(nodes) between the originator of the traffic(source)
and the receiver(destination). With this approach, the packets are routed to the
adjacent switch closet to the final destination. Then the following switches do
the same job until the packet finally arrives at its destination.

Another widely used packet-switching technique is called adaptive(or dynamic)
routing. To determine the best path between the two switches, a routing table at
the source is examined. This table is different from that of fixed directory in
which the full routing directory(contains all information about distances
between nodes) is examined at each node. In contrast, the table of adaptive
directory just contains information related to its node only. Moreover, the best
path is not determined by the shortest distance(minimum nodes between source and
destination), but by the total estimated time delay.
The Relationship of X.25 and Network Routing
X.25 does not specify the routing techniques used by the network. The
structure of the X.25 packets and the contents of the address fields in the X.25
packets often influence how the vendor determines the route of the packets to
the destination. X.25 does not use any source and destination addresses in its
datapacket. Therefore, a network that uses adaptive routing must append
addresses or routing information to the X.25 data packets for the traversal
through the network.
back to contents



Send email to: Cheung Harvey
Wai-Lun
E-mail address: mailto:[e-mail]



Return to the WPI EE535
Telecommunications and Transmission Technologies Course Homepage
window.open=NS_ActualOpen;