The History Of Layers Of Frame Relay

Published Date: 02 Nov 2017

In order to transmit data from sender station to receiver, in most general terms, we need a link through which both the stations are connected. Although, there are numerous other issues which needs to be kept in mind.

If we look back in the history, people used to talk over telephones with an operator in between to connect the lines between two users. This was possible because there were very few users. But as the demand to network grows due to its convenience, results in a substantial increase in the number of users as well. Now more and more communication path or wires are required in order to connect the users which results in huge increase in cost to setup a network.

For example if a user needs to be connected with 50 different users than it requires 50 set of communication wires, and those 50 users are connected with other 100 different users than it will require 5000 set of wires.

So as users in network increases there arose a need of such concepts that can cut the cost of hardware and still make users happy. This gave birth to the Switching technique.

Switching is done through devices known as switches. Switches are placed between the transmitting stations and connect them whenever required. Basically, switching are of two types, Circuit Switching and Packet Switching.

Circuit Switching was widely used in telephone networks to transmit voice from one end to another. To accomplish this, a dedicated path is set between end users and it remains unused by other until either of user cut off the call. So in order to increase the channel capacity and to overcome the problems in circuit switching especially in transmitting digital data over long distance, Packet Switching was developed.

In Packet Switching, data are divided into chunks know as packets and are send individually over the network. Unlike circuit switching, there is no dedicated path as now same path is shared by different users simultaneously we need to route each packet to its destination. This can be done by adding additional information to each packet which can be read by intermediary network devices such as routers (datagram approach).

Another approach of Packet Switching is of Virtual Circuit, where a pre-defined route is established between two end points and all the packets are sent through this route, so no routing is required.

As in Packet Switching, packets are transmitted through series of networking devices and links, one major issue is to resolve the conflicts between devices and network. All the network devices should be aligned with similarity so that packets can be passed through all these devices and over network safely and within time limit. Protocol, which accomplishes this task in Packet Switching, is X.25.

Origin of X.25

X.25 is very well known standard for Packet Switching and was approved by CCITT in 1976, which provides an interface between station and network. Now CCITT is known as ITU (International Telecommunication Union).

X.25 was introduced to better utilization of data channels. In circuit switching, when a connection is established, the path is dedicated and no other host can use that line. But in general, voice is not transmitted continuously. For example, when we are talking over telephone our conversation usually contains lots of small and large pauses, that is, in this pauses no data are transmitted. As a result, the channel utilization is not maximum. The idea behind the development of x.25 is to add data, which belongs to other users, between these pauses. And also X.25 can now able to deal with digital data, so the quality of data is improved.

The working of X.25 can be classified into three different levels and can be compared to the lowest three layers of OSI model as shown in the diagram[3]

CT841704.jpg

Physical Level

This level can be compared to the Physical Layer of OSI model. This level describes the issues which need to be dealt at physical layer such as connectors and cables. Besides this, physical level also,

Activate and deactivate physical circuits using electrical signals.

Maintain line characteristics of the selected interface.

Indicate faulty incoming HDLC frames, such as frames with the wrong length.

Allow configuration of auto call units (ACU) for systems with dial-up X.25 connections.

This implementation supports three physical interfaces: V.24, V.35, and X.21bis. [1]

Link Level

This level can be compared with the Data Link Layer of the OSI model and as data link layer, this level mainly deals with the frame concerns. This layer is responsible for the successful transmission of frame to the local DCE. This layer uses standard known as LABP (Link Access Protocol Balanced), which takes care of proper transmission of frames over transmission medium between DTE and DCE. It provides full duplex mode of transmission and is a subset of HDLC (High Level Data Link Control) protocol.

Each link level has a typical frame structure with header which includes address and control information and trailer which contains frame check sequence. Frames of this level can be divided into 3 categories:

Information Frame (I): Contains user data and are numbered sequentially. All the packets formed, travels in this frame.

Supervisory Frames(s): These frames are also sequentially numbered and are responsible for supervisory functions such as acknowledging, retransmission and transmission halt with respect to Information Frames.

Unnumbered Frames (U): These frames are responsible for the mode of operation, e.g. Set Asynchronous Balanced Mode (SABM).

Frame Structure of X.25 [2]

http://www.euclideanspace.com/coms/protocol/x25/link/img00011.gif

Abbreviations Used

RR: Receive Ready

RNR: Receive Not Ready

REJ: Reject

SARM: Set Asynchronous Response Mode

SABM: Set Asynchronous Balanced Mode

SABME: SABM Extended

DISC: Disconnect

UA: Unnumbered Acknowledgment Response

DM: Disconnected Mode Response

FRMR: Frame Reject Response

Packet Level

This level can be compared with the Network layer of OSI model. This layer is responsible for to set up a virtual circuit between two end DTEs and functions as a packet manager across the network.

The major working of this level includes

Setting up Call: Setting up call means to set up a virtual circuit between DTE devices and it uses X.121 addressing scheme to accomplish this task.

Data Transmission: As the name suggest, this mode is used to transmit data between two DTE devices over the virtual circuit. In this mode this level do different types of tasks including dividing and rejoining packets, error control, flow control and bit padding.

Idle: This mode is used when a virtual circuit is established but no transmission of data is going on.

Call Clear: After all the data are transmitted, the virtual circuit should be released. This mode is used to end up session between two DTEs, probably after the data transmission is over.

Restart: It is used to synchronize transmission between a DTE device and a locally connected DCE device. This mode will affect all the DTE devices involved in virtual circuits.

There could be more than one virtual circuit established between two end DTEs, so except RESTART mode, all other modes usually affect a single virtual circuit. For example, when one virtual circuit is in transmission mode then another may be in the Idle mode.

Frame Relay

Although X.25 had changed the whole scenario of data transmission from data format to link format, it has some drawbacks and these can be overcome with a better technique called Frame Relay, which focuses on the improvement of data transmission rate and frame format used in X.25.

X.25 suffers with the problem of high propagation delay between two end DTEs as there is an error checking and correction mechanism involved for data at every node it passes through virtual circuit. Although both X.25 and Frame Relay belongs to the family of HDLC, Frame Relay comparatively more efficient in LAN and for transmitting voice data and also it is easy to implement. Propagation delay is low in Frame Relay as unlike X.25 error correction is not done at every node in the circuit rather it leaves this job for end DTEs. There is an effective and logical reason behind this is that as new and new technologies are evolving, the rate of error is declining. So as error rates are falling dramatically it is time consuming to follow error correction mechanisms at every node. As a result there is no frame exchange at every Link level in between and also the three layer approach of X.25 can be completed in only two layers in Frame Relay. In other words, multiplexing and switching of logical connections takes place at layer 2 instead of layer 3, eliminating one entire layer of processing [4]. This helps in decreasing the overheads and improving the transmission rate. Moreover, in X.25, both Link layer frames and Network Layer packets requires acknowledgments. Due to this, as compared to low data rate of only 64 kbps in X.25, Frame Relay operates at a higher speed up to 1.544 Mbps.

Another improvement in Frame Relay over X.25 is difference in connection setup. X.25 setup a virtual circuit and all the data and control bits are transmitted over the same logical path whereas in Frame Relay a separate channel is used to transmit control and data bits.

Layers of Frame Relay

Physical Layer: This layer in Frame Relay is much flexible in terms of protocols. It allows implementer to use any standard protocol suggested by ANSI.

Data Link Layer: Even at this level, Frame Relay uses simple protocol as there is no requirement of flow and error control; however, it has an error detection mechanism.

Diagrammatic Comparison of X.25 and Frame Relay Working [5]

X.25

Frame Relay

Frame Format of Frame Relay [6]

http://etutorials.org/shared/images/tutorials/tutorial_157/fig286_01.jpg

DLCI (Data Link Connection Identifier): Frame Relay Header contains total 10 bits of this field and is divided into two parts. The first part is of 6 bits and second is of 4 bits.

C/R (Command/Response): This field is consist of a bit which helps upper layers to identify a frame as a command or response.

EA (Extended Address): This field also consist of a single bit and represents the status of the current byte. As it is of one bit it can have values either 0 or 1. If it is 0, another address byte is available and if it is 1 means the current byte is the final one.

FECN (Forward Explicit Congestion Notification): This one bit field value indicate the receiver about the traffic, so that the receiver can expect delay or a loss of packets.

BECN (Backward Explicit Congestion Notification): This one bit field value works just opposite to FECN, that is, it informs sender about the congestion problem in the network so that source and slow down to avoid packet losses.

DE (Discard Eligibility): This bit sets the priority of the frame. If the bit value of this field is set to 1, means network can discard the current frame in order to decrease congestion and save network from collapse.

Notes:

Data can be in any form including voice or text.

Station can be any device including a telephone or personal computer.