Communication and Networking Fundamentals

Question 1

Serial and parallel transmission

Answer 1

Serial data transmission = Where bits are sent one by one down a single wire to their destination

Parallel data transmission = Where bits are sent simultaneously down multiple wires - an 8 bit character needs 8 wires

Parallel transmission can be affected by crosstalk and skew
- Crosstalk is where the electrical signals in each wire interfere with each other; this causes data corruption and is more likely over greater distances
- Skew is where the time for each signal to reach its destination is different since each wire will have slighty different electrical properties - this gets worse over longer distances

Question 2

Serial vs parallel data transmission

Answer 2

Serial transmission:
- More reliable over greater distances since theres no chance of data skew
- No crosstalk issues
- Cheaper to make since less wires are used

Parallel transmission:
- More expensive since more cables are used
- Only reliable over very short distances
- Suffers from skew and crosstalk

Serial transmission is used over long distances (mouse to computer) while parallel transmission is used in communication between internal computer components

Question 3

Synchronous data transmission

Answer 3

Synchronous data transmission is where data is sent as one long stream - a clock signal is used to time when signals are sent and is sent along with the data itself.
This clock signal comes from the sending computer

The signals, which are sent at regular intervals, will be received in the same order that they were sent.

This makes synchronous data transmission suitable for transmitting information in real-time systems

Question 4

Asynchronous data transmission

Answer 4

Asynchronous data transmission is where the sending and receiving computer aren’t kept in sync all the time but are only synced at the time of transmission - a character is sent as soon as it is available rather than waiting for a clock signal

A start bit is used to alert the receiving computer and a stop bit is used to indicate that the transmission is over. They’re used to indicate the duration of a transmission
The stop bit is always the opposite of the start bit
The sender and receiver must use the same Baud rate and only need to synchronise their clocks during transmission

Question 5

Transmission rate

Answer 5

Bit rate = The speed at which data is transferred and is measured in bits per second (Bps)

Baud rate = The maximum number of signals changes in the medium per second and is measured in Baud, where 1 Baud is 1 signal change per second

Bit rate can be higher than baud rate if more than one bit is encoded in each signal change.

Bit rate = Baud rate × № of bits per signal

Bandwidth = The range of frequencies that a line can carry - it is directly proportional to bit rate and expressed in Hertz (Hz) (higher bandwidth = higher bit rate)

Latency = The time delay between an action and its effect - usually increases with distance

Protocol = A set of rules on how devices communicate with each other

Question 6

Physical network topology

Answer 6

Network topology = The structure of a network

Physical topology refers to how devices in a network are physically connected to each other - two types include star and bus topologies

Star topology = Network topology where each device is directly connected to a central hub - the hub recieves packets from the devices and is responsible for delivering them to the correct recipient

Bus topology = Network structure where each device is connected to a single cable called a backbone - terminators are placed at each end of the backbone

Question 7

Star vs bus topologies

Answer 7

Star topology pros:
- Packets are sent directly to their recipient, other clients can’t eavesdrop on packets that aren’t for them
- Failure of one cable doesn’t affect the rest of the network
- No collisions since each device has their own cable
- Adding and removing devices from the network is easy

Cons:
- Expensive since extra cables are needed as well as a central hub
- If the central hub fails, communication across the entire network is stopped

Bus topology pros:
- Cheaper to install since less cables are needed and there’s no central hub
- No central hub means network failure chances are reduced

Cons:
- Packets are sent through the shared backbone, which means every connected device can see packets that aren’t meant for them
- Risk of collisions since the backbone is used by multiple clients
- If the backbone fails, the entire network becomes unusable

Question 8

Logical network topology

Answer 8

Logical topology refers to how data packets flow through a network - a logical bus network delivers packets to all clients in the network and a logical star network only delivers packets to their recipients

A network set up as a star network can still behave as a logical bus network if a bus protocol is ran on the central hub

Question 9

Client-server networking

Answer 9

Host = device that provides services and resources to clients

In a client-server network, the host is a server; a powerful central computer
The clients request services from servers, who then respond with the requested service - this could be file storage, managing emails and print queues

Most schools, colleges and businesses use client-server networks to allow for central
management of clients on the network - this helps with security
Updating software and managing files is easy since they are all stored on the central server however servers are expensive to set up and take a lot of effort to maintain

Question 10

Benefits and drawbacks of client-server networks

Answer 10

Benefits:
- Generally more secure than peer-to-peer networks

• One client computer crashing does not affect the other computers
• Files and resources are easier to share and control from server
• Improved levels of security as files are centralised

Drawbacks:
- Servers are expensive

• Network technicians / specialised knowledge needed to maintain a central server
• Server failure stops the network completely

Question 11

Peer-to-peer networking

Answer 11

Peer to peer networks dont use a central server - instead, the computers act as both clients and servers - every computer has equal status and provides a different service

They are cheap to set up and maintain since there isn’t a central server however security control is limited and every computer needs to be running for the network to work
They are used when there are les than 10 users who are all in the same area

Question 12

Benefits and drawbacks of peer-to-peer networking

Answer 12

Benefits:
- Cheaper to set up, no central server needed

• Easier to set up since no specialist knowledge is needed

Drawbacks:
- If one computer fails, its files and resources cannot be accessed by the other computers in the network

Question 13

Wireless networking

Answer 13

Wireless networking allows devices to communicate within a network without being physically connected to the network

Wireless networks require a wireless access point and a wireless access adapter
The WAP connects to the wireless network and the adapter is in the device that connects to the wireless network

Question 14

WiFi

Answer 14

WiFi is used to provide wireless networks and refers to a wireless LAN that’s based on international standards - this allows a device made in one part of the world to connect to wireless networks around the world

Question 15

Wireless network security

Answer 15

Wireless networks are secured by encrypting transmitted data using WPA or WPA2
WPA = Wireless Protected Access and requires that a new wireless client enters a password to connect to the network

Another security method is disabling SSID broadcast - SSID stands for Service Set Identifier and is the name that identifies a wireless network
Disabling SSID broadcast stops wireless devices within range of the network from displaying that the network is available and only allows the devices that know the network to connect

A MAC address filter can be used to only let certain devices connect to a wireless network - every wireless device has a Media Access Control address assigned by the manufacturer
A MAC address whitelist allows specific devices to connect to a network while a MAC address blacklists blocks specific devices from connecting to a network

Question 16

The World Wide Web, the Internet and intranets

Answer 16

The World Wide Web is a system of interlinked hypertext documents that is accessed through the internet using the HTTP or HTTPS protocols to access web pages

The internet is a global network of interconnected computer networks using a globally unique address space

Intranets are private networks used only within organisations. They use internet protocols to share part of an organisation’s information with its members

Question 17

CSMA/CA

Answer 17

Carrier sense multiple access with collision avoidance (CSMA/CA) is a protocol used in wireless networks to avoid data collisions caused by multiple devices communicating simultaneously

Data is sent around networks in frames

The CSMA/CA protocol requires that:
- The sender device needs to check if the transmission channel is free
- If the channel is free the frame is transmitted; if the channel isn’t free the sender device will wait a random amount of time using a back-off mechanism before checking again - if the channel is now free, the frame is transmitted
- The reciever device transmits an acknowledgement when it recieves the frame
- If the sender doesn’t recieve an acknowledgement, it will wait a random amount of time before trying to transmit again

One problem with this protocol is that not all stations on a wireless network can be heard by the other stations - this means that a station attempting to broadcast data might not detect that another station is also trying to broadcast data, which can lead to a collision

To get around this issue, the RTS/CTS protocol is used

Question 18

CSMA/CA with RTS/CTS

Answer 18

The RTS/CTS protocol adds an extra step to the CSMA/CA protocol

The sender device sends a Request to Send signal to the WAP

• The sender device checks if the channel is free
• if the channel is free the sender sends a ‘Request to Send’ signal to the WAP; if the WAP is busy the sender device will wait a random amount of time before checking again - if it is now free the ‘Request to Send’ signal can be retransmitted
• The WAP sends a ‘Clear to Send’ signal to the sender device; when the sender recieves this signal it sends the data frame to the WAP - if the CTS signal is not recieved the sender device will wait a random amount of time before retransmitting thr RTS signal
• The reciever device transmits an acknowledgement that the data frame has been recieved
• If the sender device doesn’t recieve an acknowledgement, it will wait a random amount of time before trying to retransmit again

The WAP will respond with one CTS signal to one computer at a time - only the computer that recieves the CTS signal will transmit its data frame

CSMA/CA avoids the problem with hidden stations since only the WAP can transmit the CTS signal and the WAP can see every connected station on the network

Question 19

The internet

Answer 19

The internet is a network of interconnected computer networks that uses an end to end communication protocol - it is a wired network with underwater cables that connect different continents

An Internet Service Provider (ISP) is a company that provides its customers with access to the internet

Question 20

Packet switching and routers

Answer 20

A packet is a container in which data is transferred over the network

A packet switched network is a network where data is sent in packets - one message is split into multiple packets and are reassembled by the reciever device

Packets have to pass through multiple routers before reaching their destination; every time a packet passes through a router is called a hop

Each packet can only pass through a finite number of hops - when the packets Time To Live reaches 0, the packet is dropped (deleted).
The recipient will notice a missing packet and request the sender to retransmit the missing packet

Question 21

Packet components

Answer 21

A packet consists of:
- Sender’s address = Where the packet was sent from
- Reciever’s address = The packet’s intended recipient
- Packet contents = The data being transferred
- Time To Live (TTL) = How many hops the packet can go through before it is dropped
- Sequence number = The number of packets in a message and the packets position in relation to other packets; allows packets to be reassembled and for missing packets to be identified by the reciever

Question 22

Routers and gateways

Answer 22

Routers send packets to their recipient through the fastest possible route - either the route with the least hops or the least congested route
They connect networks that use the same protocol

Routers hold tables with information relating to the fastest route to certain devices which is updated frequently to maximise performance

Gateways strip away the packets details and leave only the packet contents - they then give the packet new sender and reciever addresses.
This is done when different networks use different protocols - the sender and reciever addresses given comply with the new network’s protocol

Question 23

Uniform Resource Locator

Answer 23

A URL is an address assigned to files on the internet - different protocols can be used in URLs to access different types of files in different ways

The URL https://www.bbc.co.uk/news/technology/index.html has many parts:
- https:// = The protocol being used to access a file - HTTPS stands for HyperText Transfer Protocol Secure
- www = Subdomain for World Wide Web - points to the server being accessed
- bbc.co.uk = The domain, bbc is the organisation name, .uk is a top level domain and .co is a second level domain
- /news = Directory of the file being requested
- /technology = Subdirectory of the file being requested
- /index - Name of the file being requested
- .html = File’s extension

/news/technology/index.html = Location of file being accessed

Question 24

Domain names

Answer 24

A domain name identifies an individual or organisation on the internet - they meed to be registered with an internet registry to ensure that they are unique

A Fully Qualified Domain Name is a domain that specifies an exact resource and can only be interpreted in one way - it will always include the server host’s name
eg. https://bbc.co.uk/news/technology/index.html is NOT a FQDN
eg. https://www.bbc.co.uk/news/technology/index.html is a FQDN since the server being accessed is specified (www - world wide web)

Question 25

IP addresses

Answer 25

An Internet Protocol address is a unique address that is assigned to every computer on the internet
Domain names have a direct relationship to IP addresses

IP addresses are made up of two parts - the first part identifies the network connected to the internet and the second part identifies a host connected to a network
eg. the IP addresses 123.123.123.45 and 123.123.123.56 identify two different hosts on the network with the IP address 123.123.123

This makes routing packets to the correct destination network easier

IPv4 addresses contain 4 sets of numbers (123.123.123.45) while IPv6 addresses contain 6 sets of numbers (123.123.123.123.45.56)
IPv6 was introduced because there are no longer enough IPv4 addresses for the number of networks connected to the internet

Question 26

Domain Name System

Answer 26

The DNS is a hierarchical naming system for computers, services or any other resource in the internet

It translates the fully qualified domain names into IP addresses associated with networking equipment for the purpose of locating and addressing these devices world wide

A DNS server is a server that translates fully qualified domain names into IP addresses - they store databases of domain names and their respective IP addresses

Question 27

Internet registry

Answer 27

An internet registry is responsible for allocating IP addresses to devices - there are only 5 registries that each operate in different parts of the world

An important part of an internet registry’s work is to protect the world’s depleting pool of unallocated IP addresses.

When a new IP address is requested, an internet registry will first look for a previously allocated IP address that has become unused rather than allocate a brand new IP address straight away

Question 28

Internet security

Answer 28

Firewalls sit between a device and the internet and regulate the packets that pass through it - they can be hardware or software and work as a proxy server that can perform both packet filtering and stateful inspection

• Packet filtering = Accepting and blocking packets based on their source IP address or the protocol they are using (which is determined by their port number)
• Stateful inspection = Examinig the contents of the packet before deciding whether to let it through the firewall or not

Proxy server = A server that sits between a public network and a private network - they manage every packet that passes between the two networks

When a device in a private network
sends a packet through a firewall
and into a public network, the
packet’s “sender” address is that of
the firewall, rather than the device’s
private IP address.
- This provides some degree of anonymity to devices on private networks as their private address is never sent beyond the private network

Question 29

Symmetric encryption and key exchange

Answer 29

Encryption is used to securely transfer data over a network

Symmetric encryption = Encryption where both the sender and reciever share the same private key which is used to both encrypt and decrypt data from both parties

Before sending any information, the sender and reciever must participate in a key exchange to ensure that they both have a copy of their shared key - this means the key can be intercepted if it is shared over a network

Question 30

Asymmetric encryption

Answer 30

Asymmetric encryption is where the two communicating devices use 4 different keys - each device has 2 mathematically related keys, one key is private and the other is public and shared over the internet

When a message is encrypted with a public key, only the corresponding private key can decrypt it and vice versa

Before a message is sent, it is encrypted by the sender using the recipients public key.
This means that the message can only be decrypted by the corresponding private key - the recipients private key.
This means that only the recipient can decrypt the message

Question 31

Digital signatures

Answer 31

Digital signatures are used in asymmetric encryption
They are used to verify the sender of a message and to verify that a message hasn’t been tampered with during transmission

1) A digest of the message is created by a checksum algorithm - the value of the digest depends on the contents of the message

2) The digest is encrypted with the senders private key - this can be decrypted with the sender’s public key

3) The encrypted digest is appended onto the message

4) The message and appended digest are encrypted with the recipients private key, meaning that only the recipient can decrypt the information

When the reciever recieves the information, it decrypts the message using its public key - this leaves the digest encrypted
The senders public key is used to decrypt the digest - this verifies that the message was really sent by the sender since only the sender has access to its private key that was used to encrypt the digest

The recipient then carries out the same checksum algorithm to see if its result matched the decrypted digest - if everything matches, the message is real and wasn’t tampered with during transmission

Question 32

Digital certificates

Answer 32

A digital certificate verifies ownership of a key-value pair used in asymmetric encryption and can be used to check that a fake key pair isnt being used by an imposter

They are issued by certificate authorities and contain:
- Serial number
- Owner’s name
- Expiry date
- Owner’s public key
- Certificate authority’s digital signature

Question 33

Worms, trojans and viruses

Answer 33

Worms are pieces of malware that can self-replicate between computers either within a network or by users downloading and running a malicious file

Trojans are a type of malware disguised as a harmless file that users can be tricked into opening - they are usually spread through emails or downloaded

Viruses are a type of malware that reside in a host file - when this file is opened, the virus is activated
They spread between computers in a private network and the internet

Question 34

Preventing malware

Answer 34

Good code quality is important in preventing malware since malware often exploits bugs in code to affect computer systems

Antivirus software also helps prevent malware - they scan the files of a computer and can remove files that contain or are malware

Question 35

Transmission Control Protocol/Internet Protocol

Answer 35

The TCP/IP protocol is used in all parts of the internet to enable different devices to communicate - it is formed by 4 different layers that form the TCP/IP stack; each one is responsible for a different part of communication over the internet

• Application layer = The highest layer and the closest one to the user. It selects and uses the correct protocol to transmit data and interacts with the user with application software like a web browser
  Gives applications access to the network
• Transport layer = Establishes a virtual path between the sender and the receiver. The layer splits data from the application layer into packets - each packet is given a sequence number and a port number which identifies what protocol is being used
• Network layer = Provides the correct IP addresses to the packets source and destination
  Routers work within the network layer by using the packet’s IP addresses to send it to its destination
• Link layer = Deals with the physical connections between the devices.
  It adds MAC addresses to the packets it receives from the network layer - these identify the hardware to which a packet should be sent; they change with every hop through a network

Question 36

TCP/IP: Recieving end

Answer 36

Once a packet has been received by its intended recipient, it is stripped of its extra information by reversing the TCP / IP stack.

Firstly, the link layer removes MAC addresses from the packet.

Next, the network layer removes IP addresses before the transport layer uses the packet’s port number to determine the correct application to send the packet to.

The transport layer also uses the
packet’s sequence number to ensure that it is in the correct position relative to other
packets in the same transmission.

Finally, the application layer receives the packets and displays the information to the user accordingly.

Question 37

Sockets

Answer 37

A socket is the combination of the host IP address and port number

The connection between 2 sockets provides a two-way communication path between the end processes

Question 38

MAC address role

Answer 38

A MAC address is a unique code that’s hardcoded onto a network card when it is made

The MAC address of a packet is read and compared to the network adapters MAC address - if they match the network adapter passes the packet to the link layer of the TCP/IP stack

Question 39

How the protocol stack works

Answer 39

The client requests a service from a server - could be a web page or sending an email

The application layer selects the appropriate protocol for the type of communication being used - interacts with the user through application software

The transport layer establishes a connection between the user and the server - the requests are split into packets and attaches the packet number, client number and the well known port number to the packet

The network layer adds the source and destination IP addresses to the packet

The link layer adds the source and destination MAC addresses to the packet

When the network layer recieves packets from the link layer it strips off the IP addresses from the packet

When the transport layer recieves packets from the network layer it reassembles the packets and uses the port numbers to route the assembled packets to the correct application layer server

The server uses the recieved IP addresses to know where to send ita response to

Question 40

Ports

Answer 40

Port = 16bit value used to identify applications and services on the internet

Well-known port numbers are standardised port numbers used to identify specifc services

Client port numbers are port numbers that only exist for the duration of the connection between a client and a server

Question 41

Application level protocols

Answer 41

• FTP = File Transfer Protocol [Port 21 and 20] = Used for sending files between devices; one device runs an FTP client while the other runs an FTP server
  Most FTP servers restrict access and require a username and password, which is known as non-anonymous access
  Port 21 is for commands and port 20 is for data
• HTTP = HyperText Transfer Protocol [Port 80] = A client sends a request message and the server responds with a response message - usually a web page
  HTTPS does the same thing but encrypts the data during transmission
• Web servers hold pages in text form which is delivered to clients through HTTP
  Web browsers receive these text files and render them as web pages
• POP3 = Post Office Protocol 3 [Port 110] = Used for retrieving emails from an email server; it communicates with email server to check for and download any new messages
• SSH = Secure Shell [Port 22] = Used for remote management and encrypted communication over a TCP/IP network
  SSH uses public key encryption to authenticate users - The SSH client and server software encrypt messages before they are sent through a TCP connection
• Using SSH with other application level protocols means you can create a tunnel through port 22 that HTTP, POP3 and SMTP requests can travel through; this means that an HTTP get request can be sent securely through SSH encryption but can also bypass network restrictions that may have been placed on other ports
• SMTP = Simple Mail Transfer Protocol = Used for sending emails and is used between a client and an email server

Question 42

Subnet masking and IP address structure

Answer 42

An IP address is split into two parts: a network identifier and a host identifier.
Each of the computers in a network shares the same network identifier but has its own host identifier.

Networks can be divided into subnets, which are smaller networks that each have a different network identifier
This network identifier part of an IP address can be found with a subnet mask

To find the network identifier part of an IP address, the device’s IP address is ANDed with the subnet mask:
1) Change all the values of the devices IP address and the subnet mask to their binary equivalents
2) AND each bit together
3) Convert the AND result back into decimal to get the network identifier

The more bits that are assigned to the network identifier of an IP address, the more different subnets a network can have.
In the same way, the more bits that are assigned to the host identifier, the more different devices can be connected to each subnet simultaneously

Question 43

Public and private IP addresses

Answer 43

Routable IP addresses are public while non-routable IP addresses are private IP addresses

IP addresses define where a host is on the internet

Routable IP addresses are unique and used on the internet - houses can share 1 IP address but each wireless device in the house will have different non-routable IP addresses

Routable IP addresses are unique while many devices can share the same non-routable IP address as long as they are on different networks

Question 44

Dynamic Host Configuration Protocol

Answer 44

DHCP is used to assign IP addresses to devices as they join a network

DHCP uses a pool of available IP addresses to allocate IP addresses to new devices for the duration of their session
Once that device leaves the network, its IP address is returned to the pool of available IP addresses

This is done since the number of available IP addresses in a private network is limited

Question 45

Network Address Translation

Answer 45

NAT is used to convert IP addresses as they pass through a router connecting a public address and a LAN with a private address space

When a device on a private network needs to communicate with a device on the Internet, it sends packets through the router, which makes a record of the packet before replacing the private IP address of the computer with its own routable IP address.

When a response is received, it is sent to the router’s public IP address, which then forwards the response to the correct private IP
address by using the record it made when sending the packet

Question 46

Port forwarding

Answer 46

Port forwarding is used when a client needs to communicate with a server in a private network

The client sends packets to the public IP address of the router belonging to the server’s private network.

The packets sent by the client contain the port number of the application running on the server that the client wishes to access.

The private network’s router then forwards the packets to the server using NAT

Question 47

The client server model

Answer 47

A client server network is one where resources, security and other functions are centrally controlled by the server - this model is used on the internet

A client running a web browser can request a web page from a web server through HTTP - the web server uses HTTP to send a response back to the client

Question 48

Application Programming Interfaces

Answer 48

An API is a set of routines and protocols that relate to how different applications communicate with each other

APIs separate the data from the operations that may be performed on the data - this makes it possible for applications to share data and take actions on one anothers behalf without needing developers to share code

The websocket protocol is an API that operates in the application layer of the TCP/IP stack - it can provide a constant stream of information between two devices, usually a web browser and a server

The connection created is duplex; data can be sent in both directions at the same time
- Websocket packets are also smaller since the size of packet headers is reduced

This is used in video streaming and instant messaging

Question 49

CRUD and REST

Answer 49

CRUD stands for Create, Retrieve, Update, Delete - these commands are used to query online databases

CRUD = SQL:
- Create = INSERT
- Retrieve = SELECT
- Update = UPDATE
- Delete = DELETE

REST is an acronym of Representational State Transfer - it uses the HTTP request methods to query databases They are GET, POST, PUT and DELETE

HTTP = SQL
- GET = SELECT
- POST = INSERT
- PUT = UPDATE
- DELETE = DELETE

Question 50

Connecting a client to an online database

Answer 50

When a client needs to connect to a database, it follows these steps:

1. Client-server request made by the client to the web browser
2. Web browser responds with the requested web page (which is delivered as a text
   file)
3. This text file contains JavaScript which loads an API. The API uses REST to enable
   the database server to be queried by the client with the use of HTTP request
   methods.
4. The client sends HTTP requests to the database server
5. The database server responds to the client’s requests using either JSON or XML.
6. The client’s browser processes the JSON or XML and displays the response to the user

Question 51

XML and JSON

Answer 51

Database servers respond to queries using XML or JSON

XML = Extension Markup Language

JSON = JavaScript Object Notation

JSON is more compact, easier to read and create, and faster for computers to process compared to XML

XML is sometimes more flexible than JSON

Question 52

Thin client computing

Answer 52

In thin-client networks, the majority of the network’s processing power belongs to servers which provide services and resources including storage and processing.

It’s easy to add new clients to thin-client networks and the clients themselves are inexpensive machines.

Thin-client networks also allow for greater centralised control of the
network as software updates and security can be managed from the server.
However, thin-client networks require a powerful server which is expensive and requires
expertise to set up and maintain

Question 53

Thick client computing

Answer 53

In a thick-client network, the clients are powerful enough to provide their own processing power and storage.
This independence eliminates the requirement for a server, although it
is possible for thick-client networks can make use of a server.

Thick-client networks require more powerful clients than their thin-client counterparts, making the network expensive to set up.
However, the cost and expertise required in setting up and maintaining an expensive server is done away with.

Thick-client networks are harder to maintain because there is no facility to issue updates and manage security from a central server.

Compared to thin-client networks which suffer from high volumes of traffic communicating between clients and the server, thick-client networks boast much quieter communication
channels which reduces the likelihood of data collisions