To connect two or more network together is called inter networking. If two network are inter connected than they must use same protocol.

For inter networking following devices are used:

1. Repeater.

2. Bridge.

3. Router.

4. Gateway.

5. Protocol Converter.

Routing is a process of selecting appropriate route for transferring data. The mechanism inside the router by which routing process is done is called routing algorithm.

Routing is classified into two parts:

1. Adaptive Routing.

2. Non Adaptive Routing.

📌Highlights: Hi friends ! In this post we will discuss about Congestion Control Algorithms such as, Pre-allocation of buffer, Scheme of packet discarding, Flow control, Congestion control by token bucket / Isarithm control, Congestion control by choke packet, Congestion control by leaky bucket, Token bucket.

Let we discuss about Congestion Control Algorithms which is used to control the congestion.

1. Pre-Allocation of buffer :

It is a prevention scheme. It is for virtual circuit subnet.

📌Highlights: Hi friends, In this post we will discuss about congestion control, What is congestion control,capacity of subnet,how to handle congestion. 

If each node send message with maximum rate than subnet are not able to carry message and generate congestion (traffic problem).

For Example, If subnet has capacity to carry only 100 messages. But the incoming message will we more than 100 than congestion control is needed.

Subnet is not design in the term of maximum number of delivery.

Capacity of subnet depend on following parameters:

1. Bandwidth of communication channel.

2. Number of buffer in various IMP's .

3. Speed of processor in IMP's.(The processing capacity of the processor used in different IMP's in packet/sec)

Now the graph will be:

Let we understand the above figure suppose here 100 packet/sec will be the subnet capacity that means subnet can hold 100 packet/sec.If subnet will receive 100 packet/sec and it also deliver 100 packet/sec than we get ideal curve as show in the above figure. But if we increase delivery rate more than 100 packet/sec to subnet than congestion / traffic will increase and the curve will go towards down.

Suppose if subnet received 150 packet / sec , than it is possible that subnet will deliver only 60 packet / sec due to congestion / traffic.In this way we can say that if we increase packet rate more than subnet capacity than delivery rate will decrease and not proper channel utilisation done. In this case received curve is called real curve.

For handling the congestion:

1. Congestion Prevention.

2. Congestion Avoidance

3. Congestion Recovery

Congestion Prevention:

In this scheme once we remove congestion than it will never occur.

(It is just like vaccination taken by any child once in a life and it will protect whole life.)

Congestion Avoidance:

In this scheme we continuously sense the channel whether congestion occur or not through any parameter like (monitor) and remove it.

Congestion Recovery:

In this scheme recovery will done after congestion occur. 

This is all about the introduction of congestion control i hope it is useful for you this part-I. In part-II we will discuss about Congestion Control Algorithms.

📌Highlights: In this post we cover data link layer protocols such as One bit sliding window protocol, Go back 'n' protocol, Selective repeat protocol and also cover problems arise with these protocols, We also discuss about there solution. We explain with the help of diagrams so that it is easily understandable.

Refer Data link Protocol #1,#2,#3

Protocol #4 : One bit sliding window protocol :

Features:

1. Data traffic is duplex hence concept of piggybacking is used.

2. Channel is noisy.

3. Both sender and receiver has one buffer each.

4. Receiver has finite and limited processing speed.

Note: Why this protocols is known as sliding window?

Protocol #1 : Unrestricted Simplex Protocol  OR Unrestricted Utopia protocol.

Features:

1. Data traffic is simplex.

2. Channel is Noise free.

3. Both sander and receiver have infinite amount of buffer.

4. Both sander and receiver have infinite processing speed.

⧭ Today we will discuss about the technique to find  sequence of redundant bit in any  message. Cyclic Redundancy Checker technique  is used to detecting redundant bit. Let we  know more about CRC (Cyclic Redundancy Checker technique).

➧ Suppose network layer send data in the form of packet such as:

1110  11110 111110 1111110 : packet

This packet is than received by DLL. It is responsibility of DLL that controlling the error and send the message to network layer of receiver end.

📌Highlights: Hello Friends ! in this post we discuss about Carrier Sense Multiple Access (CSMA) / Collision Detection (CD), CSMA/CD protocol.In this post we covered about Non persistent CSMA protocol, about One Persistent CSMA protocol, about P - Persistence CSMA protocol, how channel  can detect collision.

Let P_k is the probability that exactly K attempts are required to transmit a frame.

P_{k = ( 1- e}-G ₎^k-1 _{. ( e}^-G ₎¹

_{( 1- e}-G ₎^k-1 is Probability of failure

_{( e}^-G ₎¹  is Probability of failure    

Excepted number of attempt

E = e^G

 Where: G is load, E    is exponential load

NOTE: Number of attempt will reduce if load is less.

FINITE POPULATION ALOHA  / SLOTTED ALOHA

There are N number of station are compiling for a single channel SLOTTED ALOHA channel.

G_i : It is a load shared by the station I, that means number of frame including new entry transmission transmitted by station is per frame time) / It is probability that station ( I ) will transmit in a particular slot.

S_i : Throughput shared by a station ( I ), that means probability that particular station I will transmit successful.

OR

Mean / average number of successful transmission per frame time by station (i).

S_i = G_i   (1 – G₁) (1 –G₂) ---------- (1 – G_N)        

G = G_{1 +} G_{2 +} G_{3 + ----------- +} G_N

_{ASSUME :}

G = G_{1 +} G_{2 +} G_{3 + ----------- +} G_N

Load shared by all station are equal than.

G = N . G_i

G_i = G / N

Let the throughput of the channel is S

S = S_{1 +} S_{2 +} S_{3 + ----------- +} S_N

_{ASSUME :}

S₁ = S₂₊ S_{2 +} S_{3 + ----------- +} S_N

S_i = S /N

_than

1. Delay

2. Throughput / Efficiency / Utilisation             

Delay can be compared in the case of extreme load due to some rules. Such as only one 

can be transmitted.

In PURE ALOHA delay = 0 , because when any sander make their message immediately 

they can able to send the message without waiting at time slot.

In SLOTTED ALOHA   we have to wait for one time slot.  This delay is due to rules

 ( such as only one can be transmit / we cannot transmit until previous are not completed)

PREVIOUS POST : WHAT IS ALOHA ? IN NETWORKING PART 1

Now in this post we understand about PURE ALOHA and SLOTTED ALOHA NUMERICALLY.

Some terminology is very important when we want to calculate following things regarding ALOHA.

1. PERCENTAGE OF SUCCESSFUL TRANSMISSION.

2. PERCENTAGE OF COLLISION IN CHANNEL.

3. CHANNEL UTILISATION.

4. EFFICIENCY OF ALOHA.

Before knowing about all above things we must know about following terms :

1. FRAME TIME:

Time required to transmit standard time frame.

2. CHANNEL LOAD (G) OR OFFERED LOAD :

Mean / average number of frames including new frame and re-transmitted frame per frame time.

3. THROUGHPUT (S) : (including only newly generated frame)

    Define 1: Mean number of newly generated frames per frame time.

    Define 2: mean number of successful transmission per unit time.

    S < = 1 [ This is because at a time only one frame is generated ]

There are basically three type of frame time.

A. SUCCESSFUL FRAME TIME :

    When one frame is transferred at a time.

B. COLLISION FRAME TIME :

    When at a time more than one frame is transmitted.

C. EMPTY FRAME :

    hen no any frame is transmitted.

RELATION BETWEEN S AND G:

    G : (NEW + TRANSMITTED FRAME)

    G >  1

    When S < = 1

    Therefore    S < = G

NOTE : Only after successful transmission new frame is generated.

YOU MUST REMEMBER ALL ABOVE TERMS

NOW WE WILL FIND THE EFFICIENCY OF ALOHA BY CALCULATING  PROBABILITY OF SUCCESSFUL TRANSMISSION (P) AND THROUGHPUT (S).

PURE ALOHA:

In pure ALOHA any time we can transmit the message.

Let the probability of success is P0

    S = G . P0

    Where ;    G = Channel load

                     S = Throughput

No one can transmit between two frame otherwise collision will occur.

therefore Vulnerable period= 2 frames.

Now as we know ; According to Poisson's formula (Markov chain model)

    P_n(t) =e^-λt (λt)ⁿ / !n

P_n(t) : It is the probability that in time duration.

t       : There will be exactly (n) activity

λ      : Mean /Average no of activity per unit of time.

By this formula we can find out how many activity will perform in the next T time , If we know the average number of activity per unit of time.

IN THE CASE OF  PURE ALOHA :

    P₀(2 frame time) = e^-2G (2G)⁰ / !0

Probability of success is :

    P₀ = e^-2G 

Then the throughput is :    

    S = G . P₀

     S = G .  e^-2G

Then differentiation S with respect to G and equating to zero like:    

      ds/dg=0

      dS/dG=0

    S_max = (1/2) e^{-2x 1/2}

               ^{= 1/ 2e} 

               ^{= 0.184}

    S_max = 0.184

So here only 18.4 %  is successful transmission and the other time is waste due to collision or ideal.

So here we can say that it is a poor utilisation of channel.

GRAPHICAL REPRESENTATION :

When sufficient Input is not given to channel then under loaded condition will occur.

        U = S / ( S + C + E )

        where : S is Success

                    C is Collision

                    E is Empty

In the case of overload collision will increase.

Here in 0.5 G the efficiency of ALOHA is 0.184.

SLOTTED ALOHA:

In slotted ALOHA we can transmit in time slot.

Let probability of success is P_0.

    S = G . P₀

Where :

               G = Channel load

               S = Throughput

Now as we know ; According to Poisson's formula (Markov chain model)

    P_n(t) : It is the probability that in time duration.

If  G < 1  Under loaded

 ALOHA  means Hello.

It is first professional channel.It gives the rule that how to share the channel.It is based on satellite based communication.

EXAMPLE:

LEO cover 8000 diameter of area.

Both Up link and Down link frequency are different.

By the help of above diagram we can easily understand Up link and Down link Frequency, When station A will send any message to Up means towards satellite than it use Up link frequency.

When any station B receive any message downward means from satellite to station than it uses down link frequency.

In ALOHA message is broad castes over the entire area but the message will receive  only destination station and other station are ignore it.

Message Switching 

Message Switching is also Store and Forward type of Switching.A note (usually a special computer with a number of disks) receive a message store it until the appropriate route is free than send it.

Store and forward is considered a switching technique because their is no direct link between sender and receiver of a transmission. A message is send to the node along one path than routed along another to its destination. 

VIRTUAL CIRCUIT PACKET SWITCHING

DATAGRAM & PACKET SWITCHING

Now we will discuss about Datagram Packet Switching. It is one of the type of packet switching. In Datagram Packet Switching connection less scheme is used . In this scheme different packet can send through different path ,So sequencing is necessary. When the packets are send through different path than it is duty of transport layer to rearrange the packets before reaching at destination.

CIRCUIT SWITCHING