28 FEB 2019: Welcome Mail received from ABM Russel

Lecture [01]: Mondays, 15:00 ~ 17:00, K309, Building K, Caulfield

Laboratory [17]: Fridays, 14:00 ~ 16:00, B352, Building B, Caulfield

Unit Staffs:

1. Chief Examiner: Dr Guido Tack: Guido.Tack@monash.edu

2. Lectures delivered by:

a) ABM Russel: ABM.Russel@monash.edu

Consultation Hours: Tuesday 4pm-5pm or by appointment

Consultation times will be announced in week 2, with consultations starting in week 3. Our teaching
team will offer regular consultation times. During the consultation sessions, you can ask any
questions you have about the material presented in the lectures or labs, about your assignments, or
about exam preparations. You can attend any consultation session (not just the ones by your
allocated tutor).

3. Tutors for Laboratory: Xiaoxiang Guan:

Rules for email communication with staffs:

a) Must be sent from Monash University email account.

b) Include the following in the Subject Line:

Full Name: Mohammad Atiqur Rahman

Unit Code: FIT9135

Lecture No.: 01/ Laboratory No.: 17

Discussion forum: will be used for general issues like conception and assignment clarification etc.

Weekly Activities:

a) Self-paced reading of the textbook prior to your lecture and completing in-lecture online
quizzes

Quiz of Week 2 = Prescribed reading and lecture materials of Week 1

In-lecture Online Quiz:

 Starts from Week 2 ends in Week 11.

 Students will be asked to complete online quizzes in Moodle based on the prescribed
reading and lecture materials of the previous week.
 Quizzes will be open for 10 minutes during each lecture. Mondays: 16:00 ~ 16:10.

b) Peer instructions participation: peer instruction during lecture sessions

c) Practical exercises during tutorial/lab sessions.

 d) Reading materials: will be online on Wednesday for the following week.

Laboratory: The labs start in week 1! Each week, you will get a set of tasks on a lab sheet. Your
tutors will guide you through the tasks during the labs.

Students are encouraged to reply to a post in the forums as part of the learning process.

Peer Instruction Session: Peer instruction session (Developed by Eric Mazur of Harvard University)
will be carried out using a response gathering system. You need to bring a device that capable of
web-browsing, e.g. a laptop or smartphone.

Flux.qa (Similar to Kahoot!) will be used for answering the questions in the lecture.

Week 1:

Exam Preparation: Chapter 1: 1.2, 1.3 and Chapter 2: 2.1, 2.2

Chapter 2: Application layer

Application layer is called: layer 5

These four functions —are the basic building blocks of any application:

 Data storage,
 Data access logic (Data manipulation including using SQL, e.g. DBMS software),
 Application logic (Business logic), and
 Presentation logic (presentation of data and acceptance of user command)

Five (5) fundamental application architecture:

 Host-based architecture (first network was host-based, 1960.)

o Host (mainframe) responsibility: Data storage, data access logic, application logic
and presentation logic.
o Client (terminal) responsibility:
 Capture keystrokes
 Send keystrokes to server for processing
 Accept instruction on what to display
o Pros:
 Simple architecture
 Software and data are stored in one server
 One point of control (the central server)
 Economies of scale (centralized resources)
o Cons: (2 fundamental problems)
 Hardware: Server must process all messages. Over time they become
overloaded and slow. User prioritization becomes difficult, response time
slower. Server upgradation is costlier.
 Software: Software are difficult to use, more expensive and requires higher
development time.
 Client-based architecture (Both client and server reside on the same LAN, late 1980s)
o Host (PC) responsibility: Data storage
 o Client (PC) responsibility: Data access logic, presentation logic and application logic.
o Pros: Simple architecture.
o Cons: Network becomes overloaded as all data flow from server to client.
 Client-Server architecture (Dominant)
o Client (PC) responsibility: presentation logic, application logic (entire/split)
o Server (PC, Server farm, mainframe) responsibility: data storage, data access logic,
application logic (entire/split)
o Pros: supports multi-vendor environment.
o Cons: multi-vendor compatibility and integration become an issue.
o Solution:
 Middleware: installed between application software of client and server.
 Functionality:
o Translate software from different vendors
o Manage messages transferred between client and server.
 Standard:
o Distributed Computing Environment (DCE)
o Common Object Request Broker Architecture (CORBA)
o Open Database Connectivity (ODBC) for data access logic
o Classifications:
o 1. Tier based classification:
 2-tier (most common):
 Client (PC): Presentation logic and Application logic
 Server (PC, Server farm, mainframe): Data storage, data access logic
 3-tier:
 Client (PC): Presentation logic.
 Application Server (PC, Server farm, mainframe): Application logic.
 Database Server (PC, Server farm, mainframe): Data storage, data
access logic.
 n-tier:
 Client (PC): Presentation logic
 Web Server (PC, Server farm, mainframe): Application logic
 Application Server (PC, Server farm): Application logic
 Database Server (PC, Server farm): Data storage, data access logic
 Pros:
o Load balancing
o Scalability
 Cons:
o Network overload (Generates more Network traffic)
o Difficult to program and test software (more devices need
to communicate)
o 2. Client-load based classification:
 Thick client (fat client): All or almost all application logic on client.
 Cons: difficult to manage (update hundreds of clients along with
server)
 Thin client: Little or no application logic on client
 Pros: easier to manage (update only server NOT client)
 Impact:
 o
Thin client is the future.
o
Facilitate distributed computing model (Web browser to be
used as client software, with Javascriptor AJAX downloaded
as needed).
o Enabler of cloud-base architecture.
 Cloud-based architecture (provides scalability and rapid deployability)
 Peer-to-peer architecture

Week 2:

Study Materials:

Lecture:

Chapter 2: Application Architectures, World Wide Web

Lab: Basics of the HTTP Protocol

Week 3: Chapter 3. P57 – P81 [24 Pages]

Week 7: Chapter 5. P189 – P204 [18 Pages]

STA: Secure Ticket Authority.

In IEEE 802.11 (Wi-Fi) terminology, a station (abbreviated as STA) is a device that has the capability
to use the 802.11 protocol. For example, a station may be a laptop, a desktop PC, PDA, access point
or Wi-Fi phone.

SSID: Service Set Identifier.

BSS: Basic Service Set

BSSID: Basic Service Set Identifier

ESSID: Extended Service Set Identifier (AP with hands-Off/Roaming)

IFS: Inter-Frame Spacing

DIFS: Distributed IFS (Sender Sends)

SIFS: Shorts IFS (The time receiver waits before it sends an ACK or NAK)

RTS: Request to Send

CTS: Clear to Send

ARQ: Automatic Repeat request (Transport Layer)

WEP: Wire Equivalent Privacy

WPA: Wireless Protected Access

AES: Advanced Encryption Standard

PSK: Pre-shared Key

IEEE 802.11 MAC Frame Format [WLAN Frame (Fields length in Bytes)]
2 2 6 6 6 2 6 0-2312 4
Frame D/I Address Address Address Sequence Address Payload CRC
Control (Duration 1 2 3 Control 4
or
Connection
ID)

Frame Control: 24 bit (2 Byte)

2 2 4 1 1 1 1 1 1 1 1
Protocol Type Subtype To AP From More Retry Power More WEP Rsvd
Version AP frag Mgt Data
Comparison of Wi-Fi Types:

No. of Speed of each Max. Actual speed Frequency Technology used (for Year
Type Comments
channel channel (Mbps) range (m) (Mbps) used (GHz) generating wireless signal) released
802.11 2.4 1997
OFDM (Orthogonal Less interference.
802.11a 8 54 50 20@17m 5 Frequency Division 1999 Costly, Business
Multiplexing) clients.
Cheap, Popular, Home
users. Interference
802.11b 3 11 150 2.4 1999
with home appliances
(e.g. Microwave oven)
Higher speed, higher
coverage than
802.11g 3 54 150 2.4 OFDM 2003 predecessors.
Interference with
home appliances.
300 64-QAM
MIMO (Multiple Input
450 Multiple Output, Directing Faster, more reliable.
(with 3 Multiple streams to a single This provides a
3
802.11n antennae). Both 2.4 client). significant increase in
(1, 6, 100 2009
Wi-Fi 4 and 5 multiple data without needing
11)
600 transmitters/receivers a higher bandwidth or
(if combined all operate simultaneously at transmit power.
the channels in one or both ends of the link.
one Dual band) 4 Spatial Streams are used.
433 Signal and Data are on
300@20- 256-QAM
separate channel.
30m MU-MIMO (Multi-User
802.11ac 6900 (under 5 MU-MIMO doesn’t
8 100 MIMO, Directing Spatial 2014
Wi-Fi 5 perfect (exclusive) increase the speed to
90@Max streams to a multiple clients
condition when any single client, it
Range simultaneously)
both the AP and can increase the
 client has RTS/CTS on Separate overall data
multiple frequency range. throughput of the
antenna) Additional antenna provides entire network.
further speed. Radio beam is
Up to 1,000 shaped with special antenna
so signal is focused in certain
direction (Beamforming).
8 Spatial Streams are used
(Doubles the channel width
up to 80 MHz)
802.11ad 7000 Cannot penetrate
10 5000 2016
Wi-Gig 50,000 (future) walls
54 MHz –
802.11af 790 MHz
White Wi- (VHF and
Cognitive Radio Technology
Fi (Super UHF).
Wi-Fi) Unused TV
Channels
802.11x
2019
Wi-Fi 6
Week 12

A. The Internet model layers

B. Local Area Networks (Wired and Wireless)
C. Backbones and WANs
D. Security
E. The Internet

A. The Internet model layers.

THE FUNCTIONS AND APPLICATIONS OF THE 5 LAYERS OF TCP/IP:

1. PHYSICAL LAYER:

FUNCTIONS:

 Considered the domain of many hardware-related network design issues such as LAN and
WAN topology
 Responsible for various encoding and signaling functions that transform the data from bits
that reside within a computer/other device into signals that can be sent over the network.
 Transmit and receive data in both wired and wireless networks.
 Primarily executed in hardware by a network interface controller (NIC)

APPLICATIONS: Wired and wireless device

2. DATA LINK LAYER (MAC):

FUNCTIONS:

 LOGICAL LINK CONTROL (LLC): This layer establishes and controls logical links between local
devices on a network.
 MEDIA ACCESS CONTROL (MAC): Controls access to the network medium.
 DATA FRAMING: Responsible for the final encapsulation of higher-level messages into frames
that are sent over the network at the physical layer.
 ADDRESSING: Labels information with particular destination location where each device has
a unique number called MAC address or hardware address.
 ERROR DETECTION AND HANDLING: Handles errors that occurs at the lower level of network
stack.

APPLICATIONS: Wifi

3. NETWORK LAYER (IP):

 ROUTING: Moving data across a series of interconnected networks. Handle incoming packets
from various sources, determine their final destination, and figure out where they need to be
sent where they're supposed to be.
 LOGICAL ADDRESSING: Every device that communicates over a network has associated with
it a logical address, sometimes called a layer three address. For example, on the Internet, the
Internet Protocol (IP) is the network layer protocol and every machine has an IP address. Note
that addressing is done at the data link layer as well, but those addresses refer to local physical
devices. In contrast, logical addresses are independent of particular hardware and must be
unique across an entire internetwork.
  DATAGRAM ENCAPSULATION: Encapsulates messages received from a higher-layers by
placing them into datagrams (packets) with a network layer header.
 ERROR HANDLING AND DIAGNOSTICS: Special protocol are used to allow devices that to
exchange information about the status of the hosts on the network/device.
 FRAGMENTATION AND REASSEMBLY: Some data link layer technologies have limits on the
length of any message that can be sent, so, the network layer splits the packet and send to
data link layer. once arrived, the pieces are reassembled on the destination machine.

APPLICATIONS: IP address and IPv4 & IPv6

4. TRANSPORT LAYER (TCP/UDP)

FUNCTIONS:

 Generates the final address of the destination.

 Provides services that support reliable end-to-end communications.
 Responsible for all end-to-end communication facilities.
 Packetization of the mess, breaking up of the message into packets of reasonable size.

APPLICATIONS: TIP/VDP

5. APPLICATION LAYER

FUNCTIONS:

 This is where the message is created.

 Includes any application that provides software that can communicate with the network layer.

APPLICATIONS: Email application

Routing: 3 components

a) Routing protocol: How to find best route.

b) Routing table: Current best routes are stored in a lookup table
c) Routing decision: where to send the current incoming packet.

Routing: 2 types

a) Centralized: A central computer takes routing decision. Used in small network

b) Distributed: Individual router takes rouging decision. Routing information exchanged among
routers for populating routing table. Used in larger network e.g. the Internet. 2 Types:
a. Static routing: prepared by network manager and manually updated upon change.
Used in small network
b. Dynamic routing: Routers exchange information to populate routing table
dynamically. Initial table may be set up by network managers. 2 algorithms:
i. Distance vector: Routers exchange information about the distance to
destination. Choose shortest path (least hop count). Example:
 RIP: All routers send their routing tables to all other routers. Table
converges after some time.
a. Features: Distance = Hop count
  Max hop count: 15
 Avoids loops
b. Limitation: Only used in small networks
 Max 15 hops
 Transmits whole table during update. Creates network
overload
 Can be slow to converge
 , EIGRP, BGP
ii. Link state: Routers exchange information about the quality of links. Choose
fastest path. Example: OSPF
1. Connectivity: Routers exchange information about connectivity
 NOT just routing table (best routes) but all of the network it
knows about.
 Use a metric (usually link speed) to describe the quality of
each connection.
2. Topological map: Each router creates a topological map based on
the information received from other routers.
 a map of the entire network
 each router can independently compute best route to every
subnet using a shortest path first algorithm.
i. Dijkstra’s shortest path algorithm:
 Tentative distance: 0
 Metric: the lower the better.
 Each router calculates the distance and
finally produce a shortest path first (SPF)
tree.
 OSPF:
i. Widely used in large enterprise network.
 V1(1987), V2 (1991/1998), V3 for IPV6
(1999)
ii. It is a link-state routing protocol
 Router builds and maintains link-state
database (LSDB)
 Link cost (metric) based on Bandwidth (link
speed)
 Router uses multicast and unicast messages
to exchange LSDB information
 LSDB used to create SPF tree.
 Best routes from SPF tree are placed in
routing table.

Routing on the Internet: Autonomous Systems (AS):

 Networks operated by a single organization

 Interior routing is used for routing packets within AS (RIP, EIGRP, OSPF)
 Exterior routing is used for routing packets between different AS (BGP)
Application Layer:

HTTP is stateless meaning each request is an independent transaction

To add state to http: 2 approaches

 Client sends a session identifier with every request

 Server sends a cookie (client sends this cookie for every future transaction)

Email Server:

SMTP: for transferring text data.

MIME (Multi-purpose Internet Mail Extension): for sending attachment.

Physical Layer:

Transmission:

 Digital Transmission of Digital Data: Encoding/Decoding [RZ, NRZ, NRZI, Manchester]:

Ethernet
 Analog Transmission of Digital Data: Modulation/Demodulation [Modem]: ADSL, WiFi
 Digital Transmission of Analog Data: Sampling, Nyquest theorem N>2fh Coder/Decoder
[CODEC]: Digital Telephone Network

Digital Transmission:

 RZ (Return to Zero): Return to 0 after each bit.

 NRZ (Non-Return to Zero): 0 (H), 1 (L)
 NRZI (Non-Return to Zero Inverted): 0 if no transition (0->0 or 1->1), 1 if transition is found
(0->1 or 1->0)
 Manchester encoding: if o is found, H->L (in the middle), if 1 is found L->H (in the middle).
Ethernet used it.

Data Link Layer:

LLC (Logical Link Control):

 Interface with Network layer

 Error control
 Controls PDU header and trailer

MAC (Media Access Control):

 Error detection
 Encoding/decoding of symbol and bits between physical layer symbols and bits (frames)
  Controls when the device transmits

When to transmit:

 Roll call polling

 Token passing/Hub polling

Transmission technique:

 Asynchronous Transmission: One character (byte) [7-bit ASCII] is sent independently

 Synchronous Transmission: Several bytes [1500 bytes] sent together in a frame [SDLC
(Synchronous Data Link Control), HDLC (High-level Data Link Control), Ethernet, PPP]

FCS: Frame Check Sequence

ADSL

Voice: 4 kHz [0-4 kHz] (4-25kHz guard band)

Upstream: 113 kHz [25-138 kHz]

Downstream: 966 kHz [138-1104 kHz]

Error Detection: 3 Techniques

 Parity Check:
 Checksum:
 Cyclic Redundancy Check (CRC): P%G = R. P and R are sent. Receiver does P%G = R OK.

Error Correction:

 Backward Error Correction: Resend the message

 Forward Error Correction: Receiver repairs the frame. Used if retransmit is not possible or
higher propagation delay. Detect error-locate erroneous bit-revert erroneous bit to repair.
Example: Hamming Code, Reed Solomon code (CD/DVD, Digital TV, ADSL)

ARQ (Automatic Repeat Request):

 Stop and wait ARQ: Sender waits for ACK after sending every packet
 Continuous ARQ: Sender continuously sends packet. When receives a NAK then resends
packet

TCP Layer:

Segmenting and ARQ. Ports. End-to-end channel

Dynamic IPv6 Addressing:

 Link-local address: Special net prefix + MAC address

 Sending Multicast to all routers in LAN: Routers advertise with their subnet prefix
 Auto-configure: Advertised prefix + Self-generated Interface ID.

DNS Servers: Distribute database. Organized in a hierarchy reflecting the structure of the domain
name. Iterative (recursive) DNS lookup.
ARP (Address Resolution Protocol): Data Link Layer

TCP:

 Connection oriented – virtual point to point full duplex ckt is established between two
devices
 Reliable - Errors are detected and corrected. Segments are reassembled in correct order
 Used by: http, smtp, imap, ssh

If TCP sends big segments IP requires fragmentation

IP Fragmentation: IP fragmentation is an Internet Protocol (IP) process that breaks packets into
smaller pieces (fragments), so that the resulting pieces can pass through a link with a smaller
maximum transmission unit (MTU) than the original packet size. The fragments are reassembled by
the receiving host.

TCP Segment size: 2 approaches

 Use reasonable MTU size

 Use Path MTU Discovery (PMTUD): send IP packets to destination, asking routers to never
fragment. increase packet length until error occurs, then use last known error-free MTU.

How fast to send segments:

 Receiver transmits its maximum buffer size

 Sender sends segments without waiting for ACK up to buffer size

TCP Header: size 20 Byte

Sequence Number: Number of first data byte

Acknowledge Number: Number of data byte that I receive.

TCP Three-way handshake:

Four-way handshake during termination of the link:

Client -----Fin

---------Ack Server

----------Fin Server

Client -----Ack

UDP: 8 byte packet

 Connectionless – no virtual ckt

 Small header
 Use: DNS, DHCP, RTP

VoIP:

 Guaranteed transport
 Low latency
 Low jitter

QOS:

 Integrated Service: Application requests a channel with guarantees. Implemented by

ReSource reserVation Protocol (RSVP) on top of UDP.
 Differentiated Service: Individual packet requests a certain service class which is given
priority. Implemented by special field in IP header.

B. Local Area Networks (Wired and Wireless)

Wired LAN:

LAN Types:

Dedicated Server LAN: Business/office LAN. Server-Client Topology. Different server-Different

Function

Peer to Peer LAN: Home LAN. All computers are both server and client.

NOS (Network Operating System)

 Server NOS: NM tools, Network Service (TCP/IP protocol suite, Application Layer protocols
[http, smtp, ftp, ssh, telnet, LDAP, SQL database etc.] Replaces or Augments normal OS.
example: Linux, MS Windows Server 2012)
 Client NOS: Network Access(all layers). Part or OS (Windows 10, Mac OS, Linux, Android etc.)
Services provided by NOS:

 Directory Service (LDAP – Lightweight Directory Service Protocol. e.g. Active Directory. LDAP
contains: user[name, organization, access right, password, photo], hardware [IP Address,
Network Mask, physical location, purchase date, maintenance interval] and software
[license, access right] information)
 Network and User profile
 File, web, print, email server

Ethernet: 802.3

Developed: 1973, Standardized: 1980

Hub Based Ethernet: Physical topology: Star (Point to point), Logical topology: Shared BUS
(Multipoint). Half-duplex. Runs 50% of capacity.

Switched Ethernet: Logical Star. Full diplex. Media not shared. No broadcast. Reads MAC address.
Send to the destination only. Uses Forwarding Table. When this table is empty switch acts like a hub.
Sends the packet to everyone except the sender. Runs at 95% of capacity

Max Distance: 100m (hub-hub or client-hub)

Contention-based media access.

Switch operation modes:

 Cut-through: transmits immediately as soon as DA has been read from the ethernet frame.
Low latency. Potential for transmitting packet with error
 Store & Forward: Gets the entire packet, checks for error then transmits. High latency. Low
error. Low retransmit.
 Fragment-free: Reads header only (first 64 bytes) and transmit. Compromise of the above
two.

Reducing Network Demand:

 Move files to clients (Heavily used software)

 Install disk caching software (caches heavily used files)
 Backups at off-peak hour

Wireless: BSS, ESS, ISS

Hidden Node Problem: When a computer transmits at the same time because it cannot sense that
another computer on the WLAN is currently transmitting (Because it is out of range of the WiFi
signal of the other computer NOT the AP) is referred to as the Hidden Node Problem

Soln: Exponential Backoff (1st collision: 0 or 1 unit time, 2nd col.: 0 and 3 time unit, 3rd Col.: 0 and 7
time unit)

WLAN Channels:

 2.4Gz: 2.4000 – 2.4835 Ghz

 Total no. of channels: 13
 Non-overlapping channels: 1, 6, 11 (Mostly used) 14 (Used in Japan)
  Each channel’s width: 22 MHz
 Distance between two channels: 5 MHz

WLAN 4 datalink addresses:

 SA: Source Address

TA: Transmitter Address
 RA: Receiver Address
 DA: Destination Address

DS: Distribution System (LAN)

WLAN: Collision avoidance – 2 ways

 Stop-and-wait ARQ (This is additional to L4 ARQ): devices only send next frame after
receiving ACK for previous frame
 Controlled Access: Device sends RTS (Request To Send). AP sends CTS (Clear To Send).
Usually Not implemented

WLAN: Carrier Sense

IFS: Inter-Frame Spacing Time (High priority: Short IFS, Normal: IFS, Resend after Collision: Long IFS)

WLAN Topology:

 BSS: Independent BSS. Devices directly talk to each other. No AP is required

 Infrastructure BSS: Devices talk to AP
 ESS (Extended Service Set): Roaming between different AP (15% overlap). However, 802.11
does not support roaming BETWEEN different ESS. It requires higher level protocol (e.g.
Mobile IP).

WLAN Design consideration: Overlapping channel, interference (walls other SSID), coverage,
roaming

WLAN Attenuation: 3db attenuation means signal strength halved. 6 db one fourth.

WLAN Security:

 Open WLAN - No security

 WEP (Wired Equivalent Privacy) – all communication is ENCRYPTED with a SHARED key of
length 40 or 128 bits
 WPA/WPA-2 (Wi-Fi Protected Access) – Key is longer and changed for every packet, uses
strong AES encryption, two mode: PSK(Pre-shared Key) for personal use. AAA server for
enterprise.

Wi-Fi is half duplex. Wired LAN is Full Duplex.

C. Backbones and WANs

Advantages of VLAN:

 Subnets organized by functions (People can carry IP address even if they shift office to a
different building)
 Limited broadcasting: Only to the subnets not all the ports of the switch.
 Efficient: faster, cheaper and easier to config than routing.

CSU/DSU: Channel Service Unit and Data Service Unit (DTU: Data Terminal Unit. CPE: Customer
Premise Equipment.): These are Modems.

PAD: Packet Assembly Disassembly

Packet Switched Data Rate:

CIR (Committed Information Rate): DE (Discard Eligible). Packets sent at the rate exceeding CIR are
DE

MAR (Maximum Allowable Rate): Packets sent at higher than MAR will be discarded.

Carrier Ethernet: Same as LAN. No extra PAD/CSU/DSU needed. Cheaper but slower than ATM

VPN: Virtual NIC. Encapsulates and encrypts

Intranet VPN: Connects LAN of same company

Extranet VPN: Connect company to its customer

Access VPN: Access company over the Internet. Monash uses it. Special VPN gateway at Monash.
Client has software. This gives Monash IP to the student and staffs.

Teleporting IP Address: IP address proxy. e.g. Before Netflix came to AUS.

IPSec: Internet Protocol Security

ESP: Encapsulation Security Payload

VPN Advantages:

 Lower cost
 Easy setup
 Secure
 Flexible

VPN Disadvantages:

 No performance guarantee
 Overhead (encapsulation, encryption)
 Many incompatible standard

D. Security
Denial of Service (DoS)

TCP SYN flood: send TCP SYN but never send the ACK

ICMP flood: send broadcast ping messages with fake source (Prevention: Do not reply broadcast
ping request, very source address, traffic limit, detect traffic anomaly and block them.)

Firewalls: can be router, appliance or software. 2 types:

 Packet level FW (examines packet header of incoming packet and takes decision based on
defined rules, make use of acl)
 Application level FW (NGFW/DPI: Examines application layer packet contents. E.g. examines
email attachment for viruses)
 Comparison: Application level FW is computationally more expensive, more difficult to set up
and more resource intensive than packet level FW.
 DMZ: A part of a company network that is accessible from the Internet. Not heavily guarded
like company intranet. Allow extranet’s access to DNS, WWW and Mail servers.
 Configuration: 5 pre-defined chains: PREROUTE, INPUT, FORWARD, OUTPUT, POSTROUTE
 iptables -A FORWARD -i eth0 -o eth1 -j ACCEPT

Intrusion detection and prevention: Fail2Ban (Monitor system log files and analyse login attempts.
Based on failed attempts, blocks user for hours to weeks)

Encryption: Tool to ensure confidentiality and integrity. Disguise information using mathematical
rules. Based on a key for encryption and decryption.

Symmetric encryption: Both parties use same key

 DES (Data Encryption System): Developed by US Govt and IBM. 56-bit key
 3DES: 168 bit key
 AES (Advanced Encryption Standard): 128, 192 and 256-bit key
 RC4, IDEA, Blowfish

Asymmetric encryption: Sender encrypts message with receiver’s PUBLIC key. Receiver decrypts
message with his (receiver) PRIVATE key.

 PKE (Public Key Encryption): RSA (Rivest, Shamir, Adelman), X.509 (TLS): 512 to 1024 bit

Public Key cryptography: Daffie-Hellman – paint analogy

Digital Signature: Reverse PKE. Here private key is used for encryption and public key for decryption.

TLS: Transport Layer Security: Add SECURITY layer between TCP and Application. Predecessor: SSL
developed by Netscape. TLS is IEEE standard. Current version 1.2. Certificate is used to prove its
authenticity.

How TLS work: Two phases:

 TLS Handshake: Authenticate client and server

 TLS Record: send encrypted data

Summary to Security: Countermeasures

1. Physical security
2. Firewalls
 3. Encryption

E. The Internet

VDSL: Very High-Speed Digital Subscriber Line. Eliminates crosstalk. Speed is up to 100 Mbps. VDSL
vectoring aims to eliminate this by separating the lines from each other, allowing them to operate
independently so they retain signal clarity. The industry frequently refers to vectoring as being similar
to "noise cancellation" tech in headphones: by blocking out the excess noise, the line can be more
productive.

FTTN: Fibre To The Node or Fibre To The Neighbourhood. Last mile is connected from the node instead
of the DSLAM at CO. Hence distance between client modem and DSLAM decreases (less than 1 km.).
This reduces crosstalk/noise and increases speed.

HFC: Hybrid Fibre Coaxial cable. Last mile coaxial.

DOCSIS: Data Over Cable Service Interface Specification: Shared coaxial cable of cable TV Network.
Shared 10 Gbps. MAC-layer encryption. Authentication provided.

LTE (4G): 300Mbps/75Mbps. All-IP Network. Device movement at 500kmph supported.

LTE Advanced: up to 3Gbps. Operational in Australia.

NBN is based on GPON:

 Cheaper
 Reliable (No active equipment, No water damage)
 Broadcast to customer

IPv4 depletion and solution to it:

 Move to IPv6
 Use DHCP, NAT, PAT (NAT problem: private IP, Port number changes)
 Virtual hosting (Same physical web server hosts multiple instances of websites with same
address. DNS server pointing to the same IP address for several websites)
 Use private IP Address space (10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16)

IPv6 Implementation:
  Dual stack
 6to4: Auto tunnelling is used by to IPv6 devices to communicate over IPv4 Network. 6to4
router and 6to4 relay are required.

Internet structure:

 ISP
 AS
 IXP: Logical- partial mesh between peering ISPs. Physical – Star

Internet governance:

 ISOC (IAB, IETF, IESG, IRTF)

 ICANN (IANA)
 IGF
 ITU

Why does internet still work?

 Load balancing (providing service from multiple servers. Two types: DNS based (hostname
maps to multiple IP addresses), Load balancer: request routes to different server)
 Content Caching: GGC
 CDN: servers in multiple location, own high-BW network, PoP closer to the users. Akamai.

Future Internet:

All are IP based services: 4G, VoIP

IoT: Home automation, energy (smart meter, smart grid), environmental, medical, tranport sensors,