Networking

OSI MODEL

• OSI means open system interconnection.
• ISO (International Standard Organization) gives the name of OSI.
• OSI is a collection of rules.
• Open means all the rules in OSI are open for all type of computer.
• It decide which rule are necessary for communication.
• It take module for similar type of rules.
• They divide all rules in several layers.
• Each layer can communicate with its our upper layer and our lower layer.

 There are seven layers in OSI model such as:

1. Application Layer         (seventh layer)
2. Presentation Layer        (sixth layer)
3. Session Layer               (fifth layer)
4. Transport Layer            (fourth layer)
5. Network Layer             (third layer)
6. Data Link Layer            (second layer)
7. Physical Layer               (first layer)

Each layer has their own responsibilities such as:

1. Application layer: Application layer manage application level protocol.for example, suppose four application S/W are kept in application layer and some code are provide for each S/W such as:

If  system A make some doc.in word star and mail it to the system B but we know that if doc. made in word star than it will open in word star and if system B have no such knowledge than it create a problem ,So for overcome this problem  system A & B make some agreement  that they provide code for particular S/W as given in above table.When they send some data than with this data one header is also mention which show the type of S/W.

2. Presentation Layer: It has following duties such as:

• Data Encryption.
• Data Compression.
• Character Encoding. 

Data Encryption: It mean change the code into another code.By we can provide security of our message or data that no one can understand message.Here for this purpose system A and B both make some agreement about data encryption. such as:
00-No Encryption, 01-Encrypt shift-1, 10-Encrypt shift-2, 11-Encrypt shift-3.

At the receiver end like system B again Decrypt and transform the message block to its original form.

Data Compression:  Data compression reduce number of bits to be transmitted.Data compression become practically important in transmission of multimedia such as text,audio,video etc.
Character Encoding: We know that two system usually exchange information in the form of character,string,number and so on.Here also system A and B both make same agreement such as:
00-ASCII, 01-EBCD, 10-BCD 11-BCD.

3.Session Layer: It ha following duties such as:

• Esteblishment of sesion
• To provide synchronous points (check points)
  

Establishment of Session: It allows the communication between two process to take place either in Simplex/Half duplex/Full duplex.

Data in a simplex channel is always one way.An example of simplex is Television, or Radio

A half-duplex channel can send and receive, but not at the same time.Only one end transmits at a time, the other end receives.example of half-duplex is talk-back radio, and CB Radio (Citizens Band).

In full-duplex data can travel in both directions simultaneously. Full duplex allows both way communication simultaneously.Example can be a consumer which uses a cable connection to not only receive TV channels, but also the same cable to support their phone and Internet surfing.

To provide synchronous point(check point): The session layer allow a process to add check point (synchronous points) into a stream of data.

If a system is sending a file of 2000 pages it is advisable to insert check point after every 100 pages to ensure that each 100 pages unit is received and acknowledged independently.In this case if crash happen during the transmission of page 523, re transmission begin at 501 page, page 1 to 500 need not be transmitted.

4.Transport layer: It is responsible for end-to-end delivery.It has following duties.

• Connection establish and release.
• Divide the message into fixed size data block known as TPDU (Transport protocol data unit).
• End to end error control.
• End to end flow control.
• To provide quality of service (QOS).

Connection establish and release: For adding security the transport layer may create a connection between to en ports.A connection is single logical path between source and destination that is associated with all the packets in a message.Creating connection involve three steps

1. Connection Establishment
2. Data Transfer
3. Connection Release.

By conforming transmission of all packets to a single pathway, the transport layer has more control over sequencing,flow and error detection and correction.

Divide the message into fixed size data block known as TPDU: Transport layer breaks the incoming message into segments(parts) or (packets).The size of each packets are same.Suppose it is of 1KB each packet are (TPDU).Here breaking of message into packet has importance in network.By this we can send different packet through different path, when any path is busy.The main concept is the size of packets.Actually when we send any packet then it first store in first coming router.Router has buffer,the size of buffer is decided on the basis of size of packet that is a reason the size of packet is constant.   

End to end error control: Transport layer is responsible for error control.This layer performed end to end error control.Suppose A is one end and B is another end.B has one mechanism through which it can identify the receiving message is error free or not. Error may be Damage of Frame,Lost of frame and Duplication of frame etc

End to End Flow control: Problem arise when sending speed of sender is greater than receiving speed of receiver.In transport layer flow control is performed end to end .Here sending speed of any packet of message is more than receiving speed at the receiving end due to which a packet overhead is takes place at receiving end, this create a problem for removing this problem this layer provide a flow control mechanism. 

To provide quality of service(QOS): This layer provide quality full service because the receiving message will error free.
5.Network Layer:The network layer is responsible for the source to destination delivery of a packet.Possibly across multiple network links.The network layer ensure that each packet gets from its points of origin to its final destination.If two system are connected to the sane link there is usually no need for a network layer.However if the two system are attached to different network with connecting devices between the network,there is often need for the network layer to accomplish source to destination delivery.
It has following duties:

• Routing
• Congestion Control
• Inter-networking

Routing:When two system are attached to different network with connecting device (such as Routers /Gateways) than the route or path are decide for passing the message to destination.That means router can decide the path through which the message will send by some algorithm.
Congestion Control:If each node send message with maximum rate and subnet are not able to carry message and generate congestion (traffic problem). Example if subnet has capacity to carry only 100 messages.But the incoming message will be more than 100 than congestion control will needed.
Inter-networking:Independent network are connected together to create a inter networking.It is a network of network.More than one network to form a network is called inter network.
6.Data link layer:Data link layer transforms the physical layer a raw transmission facility, to a reliable link and is responsible for node to node delivery.
It has following duties:

• Framing
• Error control between two consecutive sections.       (Same as in transport layer)
•  Flow control between two consecutive sections.      (Same as in transport layer)
• Link management.
• Physical addressing.

Framing:The data link layer divide the stream of bits received from the network layer into manageable data units called Frame.

Link management:When two or more device are connected to the same link,Data link layer protocol are necessary to determine which device has control over the link at any given time.

Physical addressing:If frames are to be distributed to different system on the network the data link layer add a header to the frame to define the physical address of sender (source) and receiver (destination).

Physical Layer:The physical layer coordinate the functions required to transmit a bit stream over a transmission media.

It has following duties:

• Electrical property of signal transmitted.
• Mechanical property of transmission.

Electrical property of signal transmitted:Electrical property of signal means all property of signal which is in the form of 0&1 (low & high) ,what is the  shape of the signal ,which modulation technique is used such ASK,PSK etc all these contain electrical signals.

Mechanical property of signal: Mechanical property is a description of physical media is used such as.same color of wire between two (agreement) which type of connector is used (router/socket)(agreement) and physical property of wire etc.
-----------------------------------------------------------------------------------------------------------------

COMMUNICATION SERVICES AND DEVICES

There are following Devices which is used for communication such as:

• Telephone system.
• ISDN (Integrated service digital network).
• Cellular phone.

 Telephone System and Cellular Phone:A telephone, or phone, is a telecommunications device that permits two or more users to conduct a conversation when they are too far apart to be heard directly. A telephone converts sound, typically and most efficiently the human voice, into electronic signals suitable for transmission via cables or other transmission media over long distances, and replays such signals simultaneously in audible form to its user.The essential elements of a telephone are a microphone (transmitter) to speak into and an earphone (receiver) which reproduces the voice in a distant location. In addition, most telephones contain a ringer which produces a sound to announce an incoming telephone call, and a dial used to enter a telephone number when initiating a call to another telephone. A landline telephone is connected by a pair of wires to the telephone network, while a mobile phone, such as a cellular phone, is portable and communicates with the telephone network by radio transmissions. The public switched telephone network has many switching centers that interconnect telephones around the world for direct communication with each other. Each telephone line has an identifying telephone number. To initiate a telephone call the user enters the destination telephone's number into a dial or numeric keypad on the phone.  (Reference by wikipedia)
ISDN (Integrated service digital network): It means Integrated Service + Digital Network.
Integrated service means if we take different service in home such as telephone, tvcable etc than if we want to integrate cables into single cable than we get integrated service.
Digital network means end to end transmission must be digital.
Basically there are two type ISDN such as:

• N-ISDN (Narrow band ISDN)
• B-ISDN (Broad band ISDN) 

 ----------------------------------------------------------------------------------------------------------------------------------
MULTIPLEXING: Multiplexing is the set of technique that allows the simultaneous transmission of multiple signals across  a signal data link.In multiplexing system, n devices share the capacity of one link. such as 

Here path means Physical channel

Channel means the word channel refers to a portion of path that carrier a transmission between a given pair of devices.

There are basically three techniques show in the above diagram.

 FDM (Frequency Division Multiplexing): FDM is an analog technique that can be applied when the capacity of O/P link or bandwidth of link is greater or equal to the sum of I/P link.

In FDM signals,generated by each sending device modulate different carrier frequency.These modulated single are than combined into a single composite signal that can be transported by the link.

Carrier frequency are separated by enough band width to accommodate (providing space) the modulated single.

Multiplexer End: When these signal are come inside the MUX than these similar signals are modulated onto different carrier frequency such as (f1,f2,f3,f4). The resulting modulated signal are than combined into a composite signal that is sent out over a media link that has enough band width to accommodate it.

De-Multiplexer End: The De-multiplexer uses a series of filters to decompose the multiplex signal into individual signals.The individual signal are than passed to a De-modulator that separate them from their carrier and passes them to the waiting receivers.

It is the problem of overlapping when more than one signal are transmitted at the same time .In below fig.there are some overlapped area.In practical application this problem arise so many time, when you are talking with any person in phone than at that time some third person voice are also listening by you it is the situation of overlapping.

For removing the overlapping problem,Guard band concept is used. It is those part of wire which is not allotted to any pair is called Guard band.

Due to this process one bandwidth will increase so the cost will also increase.

TDM (Time Division Multiplexing): In TDM time slot is allotted to each device at all time.There are two way of transmission of data in TDM.

1. Synchronous TDM.
2. Asynchronous TDM.(Statistical TDM)

Synchronous TDM: Synchronous means that multiplexer allocate the same time slot to each device at all time weather or not device has any thing to transmit.

For example Time slot a is assigned to device A alone and cannot be used any other device.Each time slot come up a device has the opportunity to send a portion of its data.If device is unable to transmit or does not have data to send its time slot remain empty.

Frame:Time slot is grouped into frames.A frame consist of one complete cycle of time slot.

Note: In a system with n I/P lines each frame has at least n slots,with each slot allocated to carrying daa from a specific I/P line.

 

Interleaving:Synchronous TDM can be compared to a very fast rating switch.As switch opens in front of device that device has the opportunity to  send a specified amount of data on to the path.The switch move from device to device at constant rate and in fixed order.This process is called interleaving.

De-Mux: At receiver the de-mux decompose each frame by extracting each character in turn.As a character is remove from a frame,it is passed to appropriate receiving device.

Note:By assigning each time slot to a specific input line,we end up with the empty slot when ever not all the line are active.In the above figure only the first two frames are completely filled.Means that quarter of the capacity of the link is being wasted.

Asynchronous (Statistical) TDM: Let we first discuss about the problem with Synchronous TDM does not guarantee that the full capacity of link is used.In fact it is more likely that only a portion of the time slot is in use at a given instance,Because the time slot are fixed.Whenever a connected device is not transmitting the corresponding slot is empty and the much of the path is wasted.

asynchronous TDM is design to avoid this type of waste.In synchronous system if we have n I/P lines the frame contains a fixed number of at least n time slots.In asynchronous system if we have n I/P lines the frame contain no more than m slots where m is less than n.In this way asynchronous TDM supports the same number of I/P line as synchronous TDM with lower capacity links.The multiplexer scan the I/P lines,accept portion of data until a frame is filled and than send the frame across the link.

Case I:

Case II:

-----------------------------------------------------------------------------------------------------------
 ALOHA

It is first professional channel.It gives the rule that how to share the channel.It is based on satellite based communication.In Aloha uplink and downlink frequency are used for communication,both are different.In Aloha message is broadcasted over the entire area but the message will receive only destination station and other station are ignore it.

Characteristics of ALOHA:
1.No of stations:Two type of aloha are used.
     -With infinite number of user/stations i.e (infinite population aloha)
     - Finite number of station. i.e (finite population aloha)

2.Single channel:All the station are sharing common uplink frequency,i.e single channel is available and there    is no multiplexing.

3.Collision:Because of single channel if two or more station transmit in overlapped time then collision will occur and information of transmitting station will be lost.

4.Time:There are two possibilities to consider a time.
     -Continuous time or Pure ALOHA
     -Slotted time or slotted ALOHA

In continuous time/Pure ALOHA,any time we can transmit the message.example whenever our message is made we switch ON and send the message.

In slotted time/slotted ALOHA,we can transmit message in any time slot.example when the bell is on then only message will send.

5.Transmit Buffer:Each station has one buffer where a standard size frame is stored for transmission,until unless the stored frame is not transmitted successfully subsequently frames can not be generated.i.e There is no queue of frame in various stations.

figure.1

All station is used common channel,so if more than one channel send data at a time then collision will occur.For overcome this collision problem each station have Random number generator.Which generate random number that mean when A send message and B,C,D,... also send data at some time frame then collision will occur and again they retransmit and collision will occur if you do same thing infinity than collision will occur.So one provision is that each station generate random number.

-------------------------------------------------------------------------------------------------------------

CSMA/CD (Carrier Sense Multiple Access / Collision Detection):

• It has wire oriented channel.
• It is used to communicate in small areas.
• Single channel can share large number of users.
• Carriers means to carry some information.
• Collision will also occur but less than ALOHA.

The CSMA Protocols:
1.Non Persistent CSMA: Before transmission each and every station suppose to sense the channel and do the following steps.
Step1.If the channel is free than transmit.
Step2.If channel is busy than wait for random amount of time and go to step1.
Step3.If collision is found (because of lack of acknowledgement) than wait for random time and go to step1.

2.One Persistent CSMA: Before transmission each and every station suppose to sense the channel and do the following steps.
Step1.If the channel is free than transmit.
Step2.If the channel is busy than continuously sense the channel until it become free and than transmit.
Step3.If collision is found (because of lack of Acknowledgement) than wait for random time and go to step1.
Note it is a Greedy approach because it continuously sense the channel.

3.P-Persistence CSMA: Before transmission each and every station suppose to sense the channel and do the following steps.
Step1.If channel is free than transmit with probability (P) (Probability of success)and differ the operation by one slot with probability (1-p)(probability of delay)
Step2.If channel is busy than continuously sense the channel until it become free and than go to step1.
Step3.If collision is found (because of lack of acknowledgement)than wait for random time and goto step1.
CD (Collision Detection):Each station has a special electronic circuit that can detect the collision because of fluctuation in voltage during collision.This circuit is active only for those station which are currently transmitting.
After detection the collision, station suppose to do following steps.
Step1.Immediately stop their transmission (because transmission are unnecessary and useless) so stop transmission will improve for channel efficiency.
Step2.Send high frequency positive analog signal called Jammin signal into channel so that other station should not inter mean during collision period.

------------------------------------------------------------------------------------------------------------

Protocol Correctness:Finite State M/C:
In general a state m/c is any device that stores the status of something at a given time and can operate on I/P to change the status and /or cause an action or O/P to take place for any given change.

                                                            I/P----------ACTION-------------O/P

A computer is basically a state m/c and each m/c instruction is I/P that changes one or more states and may causes other action to take place.

• Each computer data register store a state.
• The read only memory from which a boot program is loaded stores a state.(the boot program itself is an initial state)

• The operating system is itself a state and each application that runs begin with initial state.
• A computer system can be seen as very complex set of states and each program in it as a state m/c.

So in summarize it, a state m/c can be describe as:

• An initial state or record of something store someplace.
• A set  of possible I/P events.
• A set of new states that may results from the I/P.
• A set of possible actions or O/P events that result from a new state.

So we can say that:
A finite state m/c is one that has a limited or finite number of possible states.
An infinite state m/c can be conceived but is not practical.
-------------------------------------------------------------------------------------------------------------

Protocols:
1.Unrestricted simplex protocol OR Unrealistic (Utopia) Protocol:
It has following Characteristics:

• Data traffic is simplex.
• Channel is Noise free.
• Both sander and receiver have infinite amount of buffers.
• Both sander and reviver have infinite processing speed.

2.Stop and wait protocol:(For noice-free channel):
It has following Characteristics:

• Data traffic is simplex.
• Channel is Noise free.
• Both sander and receiver have one buffer each.
• Receiver has finite and limited processing speed.

Explanation In this type of protocol there is a limited number of buffer.So sander can send data or frame one by one that means,A can not send data until B not received.Actually this data is stored in buffer at receiver end.There is only one buffer,so when the buffer is empty than A will send the next frame.But the problems that how A will judge that the buffer is empty.So for this B will send Acknowledgement to A and A can understand that buffer is empty and send the next frame.

Note:In this protocol flow control is needed.

3.Stop and wait protocol for Noisy channel.
It has following Characteristics:

• Data traffic is simplex.
• Channel is Noisy.
• Sender and receiver have one buffer each.
• Receiver have finite and limited processing speed.

Explain: In this type of protocol one timer is used.When first bit of frame issend than the timer is ON.Suppose in CaseI sender A send A0 frame to B and timer will ON,and the sender will not send the next frame to B untill A will not get the ACK from B.If Ack is received by A within time limit than A will send the next frame A1.But if time is out and A will not get the ACK within time limit than it will re-transmit the same frame in CaseII.

Some problem will arise in this protocol such as:
 1.Delayed ACK
 2.ACK is lost and damaged.

Solution of above problem.

For solving the above problem one extra bit is added in frame which is sequence bit(0 & 1).Sender send alternative sequence no. of 0 & 1.Such as 0,1,0,1,0,1,0,1..... and receiver will also wait for such sequence of bit such as 0,1,0,1,0,1,0,1.....

4.Sliding window protocol.

It has following Characteristics:

• Data traffic is Duplex hence concept of piggybacking is used.
• Channel is noisy.
• Both sender and receiver has buffer according to size of window.
• Receiver has finite and limited processing speed.

Why it is called sliding window:-

Here we have four buffer so we can transmit four frame one after another.A0-A3 when A3 is transmitted than if ACK of A0 will come than A4 frame is transmitted and in buffer A4 frame is store and A0 will remove in this way A5,A6,A7 frame are stored and A1,A2,A3 will remove from buffer after receiving the ACK of A1,A2,A3 here we take sequence bit of two bit (i.e 00,01,10,11)

concept of piggybacking says sequence number,ACK and data in single frame.

5.Go Back 'n' protocol:

• Its a type of sliding window protocol
• It has following Characteristics:
• Data traffic is Duplex hence concept of Piggybacking is used.
• Channel is noisy.
• Sender has a large no of buffers and receiver has one buffer.
• Receiver has sufficient processing speed.

Here sender has a large no of buffer and receiver has one buffer.
mean suppose 8 buffer for transmission + 1 buffer for receiver=9 buffer for each A and B
Note:
Go back-2 mean error comes in 2nd transmission.
Go back-4 mean error comes in 4nd transmission.
and so no

In this case here in 2nd frame error will occur but A will transmit 3,4,5,6,7 frame and B will discard these frame after error.When ACK of 2nd will not come with in time limit than again 2,3,4,5,6,7 frame will re transmit.

6.Selective Repeat protocol:

• Its a type of sliding window protocol
• Data traffic is Duplex and piggybacking is used.
• Channel is noisy.
• Both sender and receiver have large and equal number of buffer.
• Receiver has finite but sufficient speed.

For example A and B both have 16 buffer each total buffer is 32+32=64

here if in 2nd frame error will occur than it will store in buffer and all frame after 2nd frame are also stored in buffer because there is large no of buffer are present at receiver end.

here selective means repeat/re transmit only those frame which has error suppose in 2nd frame.

------------------------------------------------------------------------------------------------------------

Ethernet:
In 1985,the institute of electrical and electronics Engineers(IEEE) published standards for LANs.These standard start with the number 802.
Following are some family of  IEEE standard:

• 802.1 Definition
• 802.2 Rules of LLC (Flow control,error control)
• 802.3 Ethernet (CSMA/CD,MAC Sub layer use rules)
• 802.4 Token Bus (MAC Sub layer)
• 802.5 Token Ring
• 802.6 MAN

Note: DLL is further subdivided into two in OSI model.
        DLL
          |
          |---MAC (Media Access Control) How to share common Channel.
          |
          |---LLC (Logical Line Control) Contain allduties of DLL.
   
802.3 Ethernet (CSMA/CD,MAC Sub layer use rules):Already studied
802.4 Token Bus (MAC Sub layer):Token Bus protocol was designed to solve the problem arise in CSMA/CD protocol.The basic idea of Token Bus is to treat the bus as if it were a "Logical ring" this is accomplished in the following ways such as

• Each node knows the network address of its successor in the logical ring.
• A token is passed from node to node,each node passing the token to its successor on the ring.As with token ring a token looks like a message except for a header bit saying "Iam a token".
• There is only one token and only the node which has the token called the token holder can transmit.All other nodes can only receive.
• When a node becomes the toke holder by receiving the token,it has a period of time available to it during which it can transmit if it wishes.
• If the token holder does not need to transmit,it just passes the token to its successor.
• If the token holder needs to transmit,it can do so until its time interval has expired,than it must stop transmitting and send the token onward.This implies a maximum message length,with segmenting a long messages.

802.5 Token Ring:Click here "http://www.cse.iitk.ac.in/users/dheeraj/cs425/lec07.html"
------------------------------------------------------------------------------------------------------------------------------------
 Inter-networking
To connect two or more network together is called  Inter-networking.
If two network are interconnected than same protocol must use.
Inter-networking tools are:
1. Repeaters.
2. Bridges.
3. Routers.
4. Gateways.

1. Repeaters (Re-generator):

• Repeater recognize the shape of signal that mean whether signal is 0&1.
• A repeater is an electronic device that operate on only the physical layer of OSI Model.
• A repeater allow us to extend only the physical length of a network.
• During transmission if the distance is long than after some time or after certain distance the signal will be destroyed.But a repeater instant on a link receives the signal before it become too week or corrupted,regenerate the original bit pattern and put the refreshed copy back onto the link.

2. Bridges:

• Bridges operate in both the physical and the data link layer of the OSI Model.
• Bridges can divide a large network into smaller segments.
• Bridges can control the congestion.Bridges contain a logic that allow them to keep traffic for each segment separate.
• Bridges are also connect two different or same type of network.
• Why we use bridges:
• If we take /consider two LAN.
• Upload IMAGE...
• Here if we take repeter(R) than it does not recognize the address,it only send data for one LAN to another.In this case if A want to sand data to X than data is spread over whole channel.
• If suppose A send data to B than even the data is spread over LAN1 & LAN2 which is useless.But in this case if R recognize the address than efficiency of channel will increased.So here bridges are beneficial it has ability to recognize address but not R.
• There are three implementation take place in routing bridges.
• 1. Software:In this type of implementation one program is written in bridges which match the address.
• 2.Hardware:In this type of implementation a address is store permanently in bridges.It is faster scheme than S/W implementation.But not reliable.Because address are permanent and not change it.
• 3.Firmware:S/W + H/W Implementation.

3.Routers:

• Router access to network layer address.
• Router contain S/W that enable them to determine which of the several possible path between those address is best for a particular transmission.
• Router operate on physical,DLL and network layer of OSI model.
• Router relay packets among multiple interconnected networks.
• Upload Image...
• Here in the above diagram routers are used to connect two interilly different network.
• Mostly router are used in WAN.
• Router act like a station on a network.
• The simplest function of router to recieve packet from one conneced network and pass them to second connected network.
• Router are also use to conect more than one LAN's.
• Upload Image....
• In LAN no network layer is needed because there is no need to take routing decision,here we use single channel.
• LAN uses MAC address instead of IP address.
• MAC means (Media Access control).In LAN MAC address can decide who send the data and how much.
• In LAN every computer have their MAC address.
• This MAC address is only for LAN but accept LAN no one knows about MAC address.
• Here ARP(address resolution protocol) is used to convert MAC address into IP address and also IP into MAC address.IP address can be identify only IP address.

4.Gateways:

• Gateways potentially operate in all seven layer of OSI model.
• A gateway is protocol converter.
• A router by itself transfer,accept and relay packet only accross network using similar protocols.On the other hand a gateway can accept a packet formatted for one protocol and convert into a packet formate for another protocol.
• A gateway is generally a software installed with in router.
• A gateway understand the protocol used by each network linked into the router and therefore able to translate from one to another.

------------------------------------------------------------------------------------------------------------
Switching
A manner in which two end point can communicate to each other is called switching.
Type of switching:
1.Circuit switching.
2.Store and forward switching.
    a.Message
    b.Store and forward switching.
1.Circuit switching:
upload image...

• Circuit switching is a connection oriented technique.
• In circuit switching before sending message connection is establish.
• Handshaking is done before message sending.
• Here CP (control packet and CC(connection conform packet is send.
• Here CP contain ((source address & destination address where CC does not contain.

Suppose A make connection with following path like (A-V-W-X-Y-B) than A first send CP packet to that path .When connection is establish than whole path is reserved for that particular transmission.This path is reserved until the whole transmission is not completed.After that B make a CC packet and sent to A when A receive that CC packet than it confirm that B is ready to receive data.The time which is used to make circuit or connection is called circuit setup.
Ex:circuit switching was designed for voice communication.In telephone conversion,once circuit is established it remain connected for duration of the session.

2.Store and forward switching:
a.Message switching:

• It is a store and forward type of switching
• Here A node receive a message store it until the appropriate route is free than send it.Here the note means, it is a special computer with a number of hard disks.
• Store and forward is considered a switching because there is no direct link between sender and receiver of a transmission.
• A message is delivered to the node along one path than routed along another to its destination.

Upload image...
Draw Back: 1. Delay in process.
                   2. Required large storage media.
                   3. Unpopular for direct communication.

b.Packet Switching:

• In this scheme message or data are divided into packets.
• This scheme is connectionless that mean there is no need to establish connection before transferring message.
• Each packet contain not only data but also a header with control information (such as priority code and source address & Destination address),packet are send node to node over network.At each node or router there is buffer which store those packets and than router takes decision about path or route.
• Problem:Here the problem will arise that what will be the size of buffer that store the packet.
• Solution:For removing the problem of buffer size we consider a constant size of packet.If the size of buffer according to packet size.Suppose packet size is 1K than buffer size is also 1K.

Example: A good example is SMS in mobiles.
There are two popular approach to packet switching:
1.Data gram packet switching.

• This scheme is connection less.
• In this scheme different packet can send through different path.So sequencing is necessary.
• When the packet are send through different path than it is duty of transport layer to rearrange the packet before reaching at destination.
• In this scheme packet are known as data gram.
• Here every data gram contain source and destination address.

Upload Image...

2.Virtual packet switching.
In this scheme a single route is choose between sender and receiver at the beginning of session.
When the data are send all packets of the transmission travel one after another along that route.

Now Virtual circuit transmission is implemented in two formats.
a.Switched virtual circuit (SVC).
In this method a virtual circuit is created whenever it is needed and exist up to duration of message transfer.

b.Permanent virtual circuit (PVC).
This method is comparable to leased line in circuit switching.In this method the same virtual circuit is provided between two user.

Comparison of data gram and virtual circuit:
1. Addressing:In data gram each packet contain the full source and destination address, But In virtual circuit  each packet contain a short virtual circuit number (VC).
2. State information:In data gram router do not hold state information about connection, But In virtual circuit each VC required router table space per connection.
3. Routing:In data gram each packet is routed independently, But In virtual circuit router choose when VC is setup and all packets follow it.
4. Effect of router failure:In data gram only those users whose packet were queued in the router at that time will suffer, But in VC all VC passing through it will have to be aborted.
5. Quality of service:In data gram it is difficult, But In VC it is easy if enough resource is allocated in advance for each VC.
--------------------------------------------------------------------------------------------------------------
Routing
Routing is of two type
1. Non Adaptive Routing.(Static).
2. Adaptive Routing. (Dynamic).

Non Adaptive Routing.(Static):In non adaptive routing router can not change or select new route in any condition weather route is damaged or some other problem.
Adaptive Routing. (Dynamic):In Adaptive routing router can select new path according to our circumstances.It can select new path/route for each packet.Suppose if load increase/Decrease on network layer or we can say that load is change on network layer or any route is damaged than at that condition router can change/select new route to send their packet.

Non Adaptive Routing Algorithm:
1.Shortest path / Dijkstra Algorithm:By this algorithm we can find the shortest path from all available paths.It is based on graph,Network is look like a graph , routers are node and links are edges.It is actually a weighted graph because all the weight are given on edge.In this algorithm routers contain routing table.This table contain information of node regarding cost (distance)routing table are helpful to take route decision that means which one is cheapest (shortest).It is non adaptive scheme so once the table is created than we cannot change in any time because this table is based on distance and distance never change.

2.Flooding:It is simplest routing algorithm.In this algorithm no routing table is created at any router.In this algorithm each source node send the copy of same packet to their neighbor node/router.This process is called flooding.In this scheme some packet are routing among node/router.But some are reached at destination.But the optimality of channel is lost because all channel are engaged for delivering same packet.At the destination end overhead of same packet are arise.So this algorithm is useless,but this algorithm gives one advantage that it is garentee to reach packet at destination end faster than optimal path.This algorithm generate flood of duplicate packet on network.In this scheme it is garentee that packet are reached at destination but still the flow of packet will not stop because packet are alive in network,So for removing this problem we provide lifetime for each packet.
We provide lifetime for packet by two scheme such as:
a. Hope counter method:In this method each packet contain a counter that is decremented to each hope.When counter reach to zero the packet is discarded.
b. Time Stamping:In this method time is allotted to each packed that it will reach at destination in particular time.When the time is over and that packet which will not reach at destination a end will be discarded.

Adaptive Routing Algorithm:
Adaptive routing algorithm are classified as follow:
1. Isolated adaptive routing (Hot Potato) Algorithm:In this type of adaptive routing any IMP (Interface message processor) can not discuss to other IMP for taking decision about route.It take individual decision for selection of route is called Isolated adaptive routing.

2. Centralized adaptive routing algorithm:In this type of adaptive routing every IMP ask to any othorised IMP or central agency for taking routing decision is called centralized adaptive routing.When so many IMP,s are present ,among all IPM,s there is one central IMP is called as RCC (Routing control center).At RCC all IMP send information about current condition of line whether it is working or not.So that RCC have huge database about the lines.On the basis of this information RCC create a table for each IMP for particular interval that means the table is exist for a particular interval of time.These tables are dynamic not static.Those IMP,s and line which are present closer to RCC have more traffic because information of all IMP,s are pass through that line.This scheme is not reliable because if RCC fails than whole system will stop.

3.Distributed adaptive routing algorithm:In this the node receives information from its neighboring nodes and then takes the decision about which way to send the packet. The disadvantage is that if in between the the interval it receives information and sends the packet something changes then the packet may be delayed.
   a.Bellman-Forrd algorithm.
   b.Flloyd warshall / All pair shortest path algorithm.
--------------------------------------------------------------------------------------------------------------
Link State Routing Algorithm: (It is Dynamic)

• It is adaptive routing technique & distributed control.
• In link state routing algorithm the routing table is developed with the use of distributed control adaptive mechanism.
• Each and every IMP suppose to collect link information from all its neighbors in regular interval of time and based on information collected routing table is developed.

Total number of four steps are required for above operation.
step 1:Learning about Neighbor:

• When ever router is initialize or routed it has to learn that who are the neighbor of it and what are their I.P address.
• It a compress a goal by sending HELLO packet to each and every point to point line passing from it.Other side station suppose send its address as a reply of HELLO packet.

step 2:Measuring the line cost: 

• Router suppose to know the delay in each point to point link passing from it.
• This can be accomplished by sending Echo packet to all its neighbors.
• There station and neighbor suppose to immediately send back this Echo packet back to original station.
• The total time of getting back of Echo packet is measured and divided by two to get one way delay.

step 3:Building of Link State Packet:
Each IMP of the router suppose to built a packet which contain information regarding delay in its all adjoining  links.
step 4:Distribution of link state packet:Link state packet developed by each and every IMP will be distributed through their respective neighbor in regular interval of time.
-------------------------------------------------------------------------------------------------------------
Broadcast Routing Algorithm:

• Sending a packet to all destination simultaneously is called broad casting.
• It is used in the application where Host need to send messages to many other Host.For Example:Distribution of weather report,stock exchange market,TV and radio programs etc.
• Here network construction is like a spanning tree. (Spanning tree in which all node are connected but no cycle is created is called spanning tree).

-------------------------------------------------------------------------------------------------------------
Multicasting Routing Algorithm:

• The procedure of sending a message to the group of widely separate process in network is called multicasting.
• The message are send to well defined group that are numerically large in size but small as compared to the network as whole.
• Example Different processes work together in group as distributed database system.
• Multicast routing concerned with the fact that when a process join a group it must inform its host.
• It is important that router know about which of their host belong to which group as the router tell their neighbor the information propagate through the subnet.

-------------------------------------------------------------------------------------------------------------
Congestion Control Algorithms:
When sub-net are not designed in term of maximum number of delivery.When two many packets are present in sub-net and performance degrades.This situation is called congestion.
Capacity of sub-net depends on following parameters:
1. Channel capacity.
2. Number of buffers in various IMP's.
3. Speed of processor in IMP's.
For handling congestion:
1. Congestion Prevention.
2. Congestion Avoidance.
3. Congestion Recovery.
Congestion Prevention:Once we remove it will never occur.
Congestion Avoidance:In this scheme we continuously sense the channel weather congestion occur or not.
Congestion Recovery:In this scheme recovery will done after congestion occur.

Algorithm for Congestion Handling:

1. Pre-Allocation of Buffer:

• It is a prevention scheme.It is for virtual circuit sub-net.
• This scheme is based on virtual circuit that mean logical path is establish for delivering a packet in sub-net.
• In this scheme new packet is sent on the basis of buffer status of IMP.
• For example, In the case of stop and wait protocol one buffer is reserved in each IMP(because we know in stop and wait protocol both sender and receiver have one buffer each).
• In other words we can that no connection until the buffer become free in IMP.
• In the case of selective repeat protocol (we know that both sender and receiver have a large number of buffer before connection.

2. Scheme of Packet Discarding:

• This scheme is congestion avoidance.
• In case if one IMP sent data to second IMP and buffer of second IMP is full than second will discard the data packet.
• But when any HOST wants to sent data to any IMP then we discard that data at HOST if buffer of IMP is full.
• In the data discarding process actually we will get the time to free the buffer in IMP's.

Problem: Here infinite waiting will arise when we discard any message at HOST and buffer receiving IMP is full.Suppose at some time that IMP will sent some message/packet to that HOST and waiting for acknowledgement from HOST at that condition IMP is in wait for infinite time because HOST is also wait for free buffer in that IMP.

Solution:So to resolve that problem we use piggybacking where acknowledgment is sent with data in same packet.
Note: Before discarding a packet we sent a temporary packet.

3. Flow Control:

• It is an avoidance technique.
• In this scheme all IMP's are connected other respective HOST.
• Here if delivery rate of HOST is more than receiving rate of any IMP the flow control is needed.
• For this flow control IMP send a message to HOST that slow down the delivery rate of packet.

4. Congestion control by token packet:

• It is prevention scheme.
• In this scheme concept of token is implemented.
• If capacity of sub-net is 100 token packet is sent to sub-net.If all IMP are connected to HOST.
• If any  IMP has token than only it is able to receive the packet from sub-net.
• Here checking is performed between IMP and HOST.
• Through this scheme it is guarantee that if there is 100 token than 100 packets are in network.
• In this scheme problem of congestion will never occur.

Problem:
a. How we distribute the token.
b. Suppose if we send 100  token in any time and if 20 tokens packet are lost and there is no record about these packets and still we have 100 packets in subnet than congestion will occure.

Solution: Reset over all network in any particular time and release / pump new tokens.

5. Congestion Control By Leaky Bucket Algorithm:

• Here packet delivery form bucket to network is in constant rate.So congestion will not occur.
• In bucket there is buffer which store packet which is sent by HOST.
• But for some application where instantly data rate become fast than this method is not applicable.

6. Token Bucket:

• Token bucket is implemented for overcoming the problem of leaky bucket.
• Here in token bucket leaky rate is adaptive,we can change the leaky rate when it is needed,But in leaky bucket method leaky rate when it is needed.But in leaky bucket method leaky rate is constant / fixed.
• Here one counter is added with buffer.After one regular interval of time one token is generated.Bucket contain number of buffer,if any buffer have token than it receive packet from HOST.
• For Example,If 10 token is generated to buffer than 10 packet are received.In this scheme we can also send bulk amount of data because leaky rate is adaptive.

--------------------------------------------------------------------------------------------------------------
FDDI (Fiber distributed data interface):
Fiber distributed data interface (FDDI), which is an optical data communication standard used for long distance networks provides communication with fiber optic lines up to 200 kilometers at a speed of 100 megabit per second (Mbps).
FDDI has dual primary and secondary communication rings. The primary ring works alongside the network, and the secondary ring remains idle and available for backup.FDDI was later extended to FDDI-2 for long distance voice and multimedia communication. Organizations use this medium for voice and video conferences, online lectures, news and other multimedia.
[Reference By: https://www.techopedia.com/definition/5588/fiber-distributed-data-interface-fddi]

Fiber Distributed Data Interface (FDDI) is a local area network (LAN) communications protocol that is based on a basic token ring architecture. It is fast, reliable, and manageable. It is emerging as the standard alternative to slower protocols like Ethernet and 802.5 token ring.
Following are the feature of FDDI:
1. Maximum physical transmission speed :100 Mbps
2. Maximum packet size                          :4500 bytes
3.Typical maximum length of LAN cable  :100 kilometers    
4.Typical max. length between nodes        :2 kilometers
5.Maximum number of nodes per LAN    :500
Simple Token Ring:
An FDDI ring is a length of cable laid out in a closed loop. Current standards require that the ring cable be fiber optic cable. An optical signal (light) passes through the cable (around the ring) and returns to its point of origin. Whenever a station is connected to the ring, it is physically inserted into the ring so that the optical signal passes through the station (illustrated in Figure). Stations on the ring are referred to as upstream or downstream in relation to each other. The downstream neighbor station is the first station to see a transmitting station's transmission. In Figure, station A is station C's downstream neighbor and station B's upstream neighbor.
Fig_TokenRing:

Basic FDDI Ring:
The FDDI dual ring (or trunk ring) has two separate loops (rings). One ring is called the primary ring and the other is the secondary ring, as illustrated in Figure. Most sites use the secondary ring as a backup ring. The light signal within each loop of a dual ring travels in the opposite direction from the signal in the other ring; in FDDI jargon this is referred to as counter-rotating. Because the signal travels in different directions, upstream and downstream neighbors are opposite on each ring. In Figure, where station 2 is station 1's downstream neighbor on the primary ring, station 2 is the upstream neighbor on the secondary ring.
Fig_BasicFDDIRing:

[ReferenceBy:[http://techpubs.sgi.com/library/dynaweb_docs/0630/SGI_Admin/books/FDDIX_AG/sgi_html/ch01.html]
----------------------------------------------------------------------
DQDB (Distributed Queue Dual Bus )

• It is MAC area protocol.
• It follow 802.6 standard.
• It is used for Man .

DQDB is MAC sub layer protocol.IEEE standard 802.6 used for MAN.As the name suggested dual bus means two parallel buses are running through out the city with two head ends one in each bus and these ends are in opposite directions.These head end are suppose to generate the series of empty packets rate dependency upon channel capacity.
Distributed queue means station those are interested in transmission should get chance in FIFO method but there is no central agency that can make record and control transmission that who will transfer bus.The Queue is maintained in different station connected to DQDB system in distributed manner.
Example with the help of diagrams:
1.Figure1:

                                    [ReferenceBy:http://www.nptel.ac.in/courses/117105076/pdf/6.1%20Lesson%2019.pdf]
2. Figure 2:

                               [Reference By:http://www.comlab.hut.fi/opetus/Summer_School_2004/material/DQDB]

3.  Figure 3:

                   [ReferenceBy:http://www-i4.informatik.rwth-aachen.de/content/teaching/lectures/sub/datkom/WS06-07-bonn/04_WAN_4P.pdf]
--------------------------------------------------------------------------------------------------------------