Introduction

• Line configuration• Topology• Transmission mode• Categories of networks• Internetworks

Basic Concepts

Line Configuration

• Two or more comm devices attached to a link• Link is a physical communicating path to transfer data

Point – to – Point

• Dedicated link b/w two devices• Entire capacity reserved between two devices• Normally actual path line of wire but microwaves & satellites

links also possible– E.G Remote of a TV

Multi Point Configuration• Also called multidrop• More than two devices share common link• Capacity of channel shared• If devices use link simultaneously

– Spatially shared-in terms of space– Time shared-devices take turns

Topology• Way a NW is laid out physically or logically• Two or more devices connect to a link• Two or more links form a topology• Geometrical representation of relationship of all links

Possible Relationships

• Peer to Peer : Devices share link equally

– Ring

– Mesh• Primary-Secondary :One device controls traffic

& other must transmit through it

– Star

– Tree

Mesh Topology

Star Topology

Tree Topology

Bus Topology

Ring Topology

HYBRID TOPOLOGY

The Internet

• Loosely administered network of networks• Agreed procedures for access and

intercommunication• Internetworking uses gateways, routers and

firewalls• Gateways: convert data traffic from one network

format to another. They link LANs to WANs and WANs to WANs

• An Analog signal is a continuously varying electromagnetic wave. (Clock with arm)– Have infinite values– Used in early telephone systems.– Analog signals had the drawback that they

attenuate (weaken) over long distances. Needed amplifiers to boost the signals. However, amplifiers distort the signal and introduce noise.

• A Digital signal is a sequence of binary voltage pulses (0’s and 1’s).It is discrete. Have limited values normally 0 & 1– Digital transmission avoids the noise problem by

encoding the analog signal into digital form. The digitized version is then sent across the network.

Periodic & Nonperiodic Signal• Both analog and digital signal can take two forms• Periodic Signal : A signal which completes a pattern

within a measurable time frame called Period and repeats the pattern. A sine wave is the simplest Periodic signal

• The completion of one pattern is called Cycle• Period is amount of time required to complete one full

cycle• Nonperiodic Signal: Also called Non-periodic which

changes pattern over time.• In data comm periodic analog signal (use less

bandwidth) and nonperiodic digital signals (variation in data can be represented) are used

Units of Periods & FrequenciesUnits of Periods & Frequencies

UnitEquivale

ntUnit

Equivalent

Seconds (s) 1 s hertz (Hz) 1 Hz

Milliseconds (ms) 10–3 s kilohertz (KHz) 103 Hz

Microseconds (ms) 10–6 s megahertz (MHz) 106 Hz

Nanoseconds (ns) 10–9 s gigahertz (GHz) 109 Hz

Picoseconds (ps) 10–12 s terahertz (THz) 1012 Hz

Example 1Example 1

Express a period of 100 ms in microseconds, and express the corresponding frequency in kilohertz.

SolutionSolution

From Table 3.1 we find the equivalent of 1 ms. We make the following substitutions:100 ms = 100 10-3 s = 100 10-3 10 s = 105 s

Now we use the inverse relationship to find the frequency, changing hertz to kilohertz100 ms = 100 10-3 s = 10-1 s f = 1/10-1 Hz = 10 10-3 KHz = 10-2 KHz

Wave Length

• Wave length is another characteristic of signal moving through medium. Distance signal can travel in one period

• It binds period or frequency of sine wave to the propagation speed of the medium

• Frequency is in dependent of medium but wavelength depends upon both frequency and the medium

• Generally used in Optical Fiber

Wave Length

Wave length = Propagation sp × period

= Propagation / frequency

Wave length is normally measured in micrometers(microns)

Bandwidth• Range of frequencies contained in composite signal is

its BW• The bandwidth is a property of a medium: It is the

difference between the highest and the lowest frequencies that the medium can satisfactorily pass

If a periodic signal is decomposed into five sine waves with frequencies of 100,300, 500, 700, and 900 Hz, what is the bandwidth? Draw the spectrum, assuming all components have a maximum amplitude of 10 V.

SolutionSolutionB = fh - fl = 900 - 100 = 800 HzThe spectrum has only five spikes, at 100, 300, 500, 700, and 900

Bit rate & Bit Interval• Most digital signals are aperiodic thus period or

frequency is not appropriate• Bit Interval is time required to send one single bit• Bit Rate is no of bits sent per second• Example: A digital signal has a bit rate of 2000

bps. What is the duration of each bit (bit interval)• Solution: The bit interval is the inverse of the bit

rate

Bit interval = 1/bitrate =1/ 2000 s = 0.000500 s=500microsec

TRANSMISSION IMPAIRMENTTRANSMISSION IMPAIRMENT

• Signals travel through media, which are not perfectSignals travel through media, which are not perfect• The imperfection causes signal impairmentThe imperfection causes signal impairment• This means that the signal at the beginning of the medium is not the same This means that the signal at the beginning of the medium is not the same as the signal at the end of the mediumas the signal at the end of the medium

PERFORMANCEPERFORMANCEIn networking, we use the term Bandwidth in two contexts:•The first, bandwidth in hertz, refers to range of frequencies in a composite signal or the range of frequencies that a channel can pass• The second, bandwidth in bits per second, refers to the speed of bit transmission in a channel or link.

It is the measure of how fast we can send data. It is different from BW. We may have B BW but may send only T bpsExample:A network with bandwidth of 10 Mbps can pass only an average of 12,000 frames per minute with each frame carrying an average of 10,000 bits. What is the throughput of this network?SolutionWe can calculate the throughput as

Throughput

The throughput is almost one-fifth of the bandwidth in this case.

Time required for a bit to travel from source to destinationPropagation Time = Distance / Propagation SpeedExample:What is the propagation time if the distance between the two points is 12,000 km? Assume the propagation speed to be 2.4 × 108 m/s in cable.SolutionWe can calculate the propagation time as

Propagation Time

The example shows that a bit can go over the Atlantic Ocean in only 50 ms if there is a direct cable between the source and the destination.

Time required for transmission of all the bits.Transmission Time = Message size / BandwidthExample:What are the propagation time and the transmission time for a 2.5-kbyte message (an e-mail) if the bandwidth of the network is 1 Gbps? Assume that the distance between the sender and the receiver is 12,000 km and that light travels at 2.4 × 108 m/s.

Solution

Transmission Time

What are the propagation time and the transmission time for a 5-Mbyte message (an image) if the bandwidth of the network is 1 Mbps? Assume that the distance between the sender and the receiver is 12,000 km and that light travels at 2.4 × 108 m/s.

Solution

Example

Latency or delay is the time for a message to completely arrive at a destination from the time 1st bit left the source

Latency = Propagation Time + Transmission Time +Queuing Time + Processing delay

Latency

JitterVariance in delay. More prominent in real time applications

•When BW of a link is grater than BW requirement of devices, Link can be shared•It is technique that allows simultaneous transmission of multiple signal across single data link

Multiplexing

Dividing a link into channels

•MUX: Combines n lines to 1

•DEMUX: Separates back into its components

•Link refers to physical path

•Channel refers to portion of link that carries transmission

Categories of multiplexing

FDM• Analog Technique• Applied when BW of link greater than combined BW of signals to be transmitted• Each sending device modulate different CF, which in turn combined into a composite signal for transmission• CF separated by sufficient BW to accommodate modulated signal•Channels are separated by strips of unused BW called GUARDBAND that prevent signals from overlapping•To use FDM for digital signal covert it to analog signal first

WDM• WDM same as FDM except that it involves light signals• WDM is an analog multiplexing technique to combine optical signals• Designed for Fiber optic• Using Fiber optic cable for one single line wastes available bandwidth

Synchronous TDM •Digital process•Each connection occupies portion of time in a link•Fig shows only multiplexing and not switching i-e source 1 to any but fixed destination

Statistical TDM• Ensure no slot is wasted. Slots are not pre-assigned• Slots are dynamically allocated to improve BW• Unlike Sync TDM, total speed of input lines can be

greater than capacity of path• Slots can be less than devices• Mux scan the input line until slots are filled the transmits

LAN Medium

•Connected directly

•Signal constrained by Physical limit of media

OSI ModelPlease DO NOT Touch Steve’s Pet Alligator

Data Encapsulation• Data Encapsulation is the process of adding a header

to wrap the data that flows down the OSI model. • The 5 Steps of Data Encapsulation are:

1. The Application, Presentation and Session layers create DATA from users' input.

2. The Transport layer converts the DATA to SEGMENTS

3. The NW layer converts the Segments to Packets (datagram)

4. The Data Link layer converts the PACKETS to FRAMES

5. The Physical layer converts the FRAMES to BITS.

Types of Switching

A Datagram Network With 4 Switches (Routers)•All packets may take different route, arrive out of order, lost or dropped in the way

•These NWs are called connectionless NWs as Switch does not keep info about connection state, no setup or tear down phase

A virtual Circuit Network• A virtual-circuit NW is in between a circuit-switched and

datagram NW. It has some characteristics of both• It has setup, data transfer & tear down phases• Resources are allocated during setup phase as in circuit

switched NW or on demand as in datagram NWs• Data are packetized & each packet carries an address

(local jurisdiction only; add of next switch) in a header• All packets follow same path & implemented at data link

layer• Virtual circuit NW is implemented in Datalink layer,

circuit switched in Phy layer & Datagram NW in NW layer

A virtual Circuit Network

Single-bit error

Burst error of length 5

Protocols

• Can not be used in real life

High Level Data Link Control Protocol - HDLC

• High-level Data Link Control (HDLC) is a bit-oriented protocol for communication over point-to-point and multipoint links. It implements the ARQ mechanisms

• HDLC is most important Data link protocol• It is widely used• Supports both Half Duplex & Full Duplex TXn and

Both pt– to - pt and pt–to – multipoint• It has been developed by ISO and the standard is

called (ISO 3309, ISO 4335)

Frame Format

Flag Fields:Delimit frame at both ends with unique pattern 01111110 for syncSingle frame may be used to close one frame and open anotherReceiver hunts for flag sequence to synchronizeNot needed for pt-to-pt link but added for uniformity

HDLC Frame Types• Information Frames: User data & control info• Supervisory Frames: Control info only• Unnumbered Frames: System Management info

Multiple AccessWhen nodes are connected to a common link, there is a need of multiple –access protocol to coordinate access to the link.

Multiple-Access Protocols

Pure ALOHA• Original ALOHA is called pure ALOHA• After collision each node waits for random time

before resending frame. After maximum tries node give up and try later

Slotted ALOHA• Time is divided in to slots and each node must

transmit at the start of time slot• If node misses start of slot, it waits until start of

next slot• There is collision if 2 nodes transmit at start of slot

CSMA / CD• CSMA does not define procedure for a collision. That's

why it was never implemented • CSMA / CD adds method to handle collision• Node uses one of the persistent strategies • Any station can send frame, then monitors medium, if

senses collision , frames are sent again• To reduce probability of collision 2nd time, node waits , it

needs to back off• In the exponential back off method, node waits an

amount of time between 0 and 2 k × max_ propagation_ time where k is the no of attempted Transmissions

• 1st time = between 0 & 21 × max_ propagation_ time • 2nd time = between 0 & 22 × max_ propagation_ time

CSMA / CA• Why not CSMA/CD in WLAN: In CSMA / CD node is

able to receive while transmitting– If no collision occurs node receives own signal

(same energy as that of sent signal)– If collision occurs node receives own plus other node

signal (detected energy is almost double)• In wireless, sent energy looses its strength so received

signal has little energy, therefore collision may add only 5-10% of additional energy. This is not useful for effective collision detection

• Therefore we need to avoid collision as it can not be detected

• Collisions are avoided through 3 strategies; Inter frame spaces, contention window and Acknowledgements

Reservation• Node needs to make reservation before sending

data• Time divided in to intervals• In each interval reservation frame precedes data

frame in that interval• If there are N nodes , there are N reservation

minislots in reservation frame• Each minislot belongs to a node• When node needs to send data, it makes

reservation in its own minislot• Node that made reservation can send data after

the reservation frame

Polling• It works with topologies in which a device is

designated as Primary Station and others as Secondary Stations

• All data exchange through Primary even if its not the destination

• Primary controls the link, Secondary follow instructions

• Primary decides which node may transmit therefore its always the initiator

• If Primary wants to receive data, its asks Secondary if they have data – This function is called POLLING

• If Primary wants to send data, its asks Secondary to get ready – This function is called SELECTING

Token Passing• A node is authorized to send data if it receives as

special frame called TOKEN• Nodes arranged around a ring. Each node has a

predecessor and Successor• When no data around the ring, Token circulates• If node needs to send data, it waits fir the token• Node captures Token and send one or more data

frames either all data frame sent or allocated time expires

• Then it releases the Token to be captured by next waiting node (successor)

• Priority or reservation may be added to the process

IEEE STANDARDS

• In 1985, the Computer Society of the IEEE started a project, called Project 802, to set standards to enable intercommunication among equipment from a variety of manufacturers

• Project 802 is a way of specifying functions of the physical layer and the data link layer of major LAN protocols

Data Link Layer• Logical Link Control: Performs flow control, error

control and part of framing duties– LLC provides one single link control to all IEEE

standards• MAC layer provides different protocols to different LANs• Framing: LLC defines PDU similar to HDLC. Header

contains flow and error control functions• LLC Header also defines upper layer protocol at source

and dest. that uses LLC called DSAP & SSAP (not used by IP)

• Other fields including access methods have been moved to MAC layer. Physical layer is media dependent

Ethernet• The term Ethernet refers to the family of local-area

network (LAN) products covered by the IEEE 802.3 standard that defines what is commonly known as the CSMA/CD protocol

• Originally developed by Xerox in 1976 • Later extended by DEC (digital Equipment

Corporation), Intel and Xerox – Called Ethernet• Three data rates are currently defined for operation

over optical fiber and twisted-pair cables: Standard Ethernet (10 Mbps), Fast Ethernet (100 Mbps) and Gigabit Ethernet (1Gbps)

Connecting Devices

• Five kinds of connecting devices; Repeaters, Hubs, Bridges, Layer Two and Layer Three Switches

• Repeaters and Hubs operate in the 1st layer• Bridges and layer 2 switches on 1st two layers• Routers and layer 3 switches on 1st three layers

IPV4 Datagram Format

Supporting Protocols in TCP/IP at NW Layer

• ARP (Address Resolution Protocol )• RARP• ICMP• IGMP

Internet Control Message Protocol (ICMP)

•The IP protocol has no error-reporting or error-correcting mechanism. The IP protocol also lacks a mechanism for host and management queries. The Internet Control Message Protocol (ICMP) has been designed to compensate for the above two deficiencies. It is a companion to the IP protocol•ICMP is a mechanism used by hosts and routers to send notification of datagram problems back to the sender if some thing has gone wrong•It handles both control and error msg. It only report problem but not correct it•Since datagram carries the add of original sender & final recipient, it does not know the add of previous router that passed it along, therefore ICMP can send message only to source & not intermediate routers

IGMP (Internet Group Message Protocol)

• The IGMP has been designed to help a multicast router identify the hosts in a LAN that are member of a multicast group. It is a companion to the IP protocol.

• What is multicasting ?

AddressingAddressing• Each device on the Internet id identified through IP

add• An IP address is a 32-bit address• The IP addresses are unique and universal• The address space of IPv4 is2 32 or

4,294,967,296• Dotted-decimal notation

Netid & Hostid

Subnetting• Without subnetting, IP address with 2-level of hierarchy

(consist of netid and hostid) is not enough.• Consider the org. which has 2-level of hierarchy cannot have

more than one physical network.• With this scheme, the org. is limited to two levels of

hierarchy. The hosts cannot be organized into groups, and all of the hosts are at the same level. The org. has one network with many hosts.

• One solution to this problem is subnetting, the further division of a network into smaller networks called subnetworks. Network is divided into 3-subnetworks.

• In next fig., the rest of the Internet is not aware that the network is divided into 3 physical subnetworks: the three subnetworks still appear as a single network to the rest of Internet. E.g. A packet destined for host 141.14.2.21 still reaches router R1. The destination address of the IP datagram is still a class B address.

Subnetting• Subnetting is done borrowing bits form host portion.

Mask of all zero’s in the host field specify the entire network

• Number of bits borrowed from the host portion are identified by subnet mask. Example: Network 131.10.0.0, Mask for Class B 255.255.0.0. An eight bit mask will be 255.255.255.0. You can have 254 subnets and 254 hosts per subnet

• How to find Subnet add: Use bit wise AND operator• Rules:

1. Bytes in IP add that correspond to 255 in the mask will be repeated in Sub Net add

2.Bytes in IP add that correspond to 0 in the mask will change to 0 in Sub Net add

Default MasksDefault Masks

Network address can be found by applying the default mask to any address in the block (including itself). It retains the netid of the block and sets the hostid to 0s.

Private IP Addresses• Another approach to conservation of the IP address

space • IP addresses are globally unique by reserving part of

the address space for networks which are used exclusively within a single organization and which do not require IP connectivity to the Internet

• There are three ranges of addresses which have been reserved by IANA for this purpose:

TCP/IP and the OSI Model

Process to Process Comm• TPT layer provides process to process comm through

Client Server methodology– Process on local host is called Client and it gets

services from remote host called Server– OS support no of processes to be run concurrently

Addressing• Processes are delivered to transport layer add called

Port Number which are 16 bits integers from 0 – 65535

• Client process randomly chose a Port No where as Server Process are designated a predefined No called Well known ports assigned by central authority (1-1023) – User-defined ports (range 1024 or greater)

IANA(I/N Assigned No Authority) Ranges

•Registered: Not assigned or controlled by IANA. Can be registered to avoid duplication

•Dynamic: Neither controlled nor registered

Socket Add

• Each process need two identifiers – IP add & Port no called Socket• A transport layer protocol needs a pair of Socket add i.e. Client Socket

and Server Socket – Part of TCP / UDP Header

User Datagram Protocol (UDP)• UDP is called connectionless unreliable protocol.• It has no flow & error control mechanism• UDP message called user datagram• It is very simple protocol with no over heads thus

suitable for protocols that requires simple request - response comm with little flow & error control concerns

• Not used for app such a FTP that send bulk of data

UDP Header• Fixed header size of 8 bytes

• Checksum is over entire datagram. Calculation is optional

TCP

• It is reliable connection oriented protocol that creates virtual connection

• TCP Services:– Process to process comm– Stream Delivery Service– Buffered Service– Full Duplex Service– Connection Oriented Service

• TCP provides reliable, end-to-end data Txn with flow and error control. E.G Telnet, FTP, WWW, POP, IMAP, etc.

TCP Segment Format

Three-step Connection Establishment

Not actual but imaginary byte

What is Sync Flooding Attack? How cookies help in overcoming it?

Connection Termination using Three-Way Handshaking

Four - Way Handshaking Half-Close