What is the process of converting information into another acceptable form for transmission reverses this process in order to interpret the information?

Introduction

- In order to create such a worldwide network a framework must be set.

- Framework can also improve the development of new technologies to connect the network.

- Three elements of communication. A sender, a destination and a channel which is the media that provides a pathway for a message to travel.

- In order for communication to be successful sender and receiver must agree on how the message is sent/received so that it is successful.

Rules Establishment

- Protocols account for the following requirements:

1) An identified sender and receiver.

2) A common language and grammar.

3) Speed and timing of delivery.

4) Confirmation or acknowledgement requirements.

Message Encoding

- The first step to sending a message is encoding.

- Encoding is the process of converting information into another acceptable form for transmission.

- Decoding reverses this process in order to interpret this information.

- Messages sent over a network are first converted into bits by the sending host.

- Each bit is encoded into a pattern of sounds, light waves, or electrical impulses depending on the network media it will be transmitted.

- The destination host receives and decodes the signals in order to interpret the message.

Message Formatting and Encapsulation

- Each computer message is encapsulated into a specific format called a frame. 

- A frame acts as an envelope, it provides the address of the destination, address of the host.

- The frame has both source and destination in the frame and encapsulated message (see CH3F1).

Message Size

- Messages sent over a network are broken into smaller pieces to allow it to be more easily understood.

- The size of frames used over networks are very strict, frames that are too long or too short are not delivered. This will vary depending on the channel being used.

- Frame size restrictions require a a source to break its message into both the minimum and maximum frames.

- The message will be sent in separate frames with each frame having its own addressing information.

- At the destination host the message is reconstructed into the original message.

Message Timing

- There are 3 rules of engagement for message timing:

1) Access Method = When two people talk at the same time you get a collision of information, making it hard for either to understand and they must start again. In networking terms host need to know when to start sending messages and how to respond when an error occurs.

2) Flow Control = Is how much information can be sent and the speed it can be delivered. For networking source and destination hosts use flow control methods to negotiate correct timing for successful communication.

3) Response Timeout = Hosts on a network have rules that specify how long to wait for a response and what action to take is a response timeout occurs. 

Message Delivery Options

- 3 delivery options:

1) Unicast = A one-to-one delivery option. Only a single destination for the message.

2) Multicast = A one-to-many delivery option. The delivery of the same message to a group of host destinations simultaneously.

3) Broadcast = A one-to-all delivery option. If all hosts on the network need to receive the message at the same time.

- Some protocols use a special multicast message sent to all devices making it essentially the same as a broadcast.

- Additionally hosts may be required to acknowledge the receipt of some messages while not need to acknowledge others. 

Rules that Govern Communications

- A group of inter-related protocols required to perform a communication function is called a protocol suite. 

- Protocol suites are implemented by hosts and networking devices in software, hardware or both.

- Protocols exist in a stack which shows how individual protocols within a suite are implemented.

- Protocols are viewed as layers. Each layers serves the layer about it providing services focused on the content being sent.

Network Protocols

- Network protocols define a common format and set of rules for exchanging messages between devices.

- Common networking protocols include Hypertext Transfer protocol (HTTP), Transmission Control Protocol (TCP) and Internet Protocol (IP).

Protocol Interaction

- Communication between a web server and a web client is an example of interaction between several protocols (see CH3F2).

- HTTP = is an application protocol that governs the way a web server and web client interact. Sets the standard for formatting of the content requests and responses that are exchanged between the two. HTTP requires other protocols to govern how the messages are transported between the two.

- TCP = is a transport protocol that manages individual conversations. Devides the HTTP messages into smaller pieces called segments. Also responsible for controlling the size and rate messages are exchanged between the server and client.

-  IP = is responsible for taking the formatted segments from TCP and encapsulating them into packets. This includes assigning them the appropriate addresses and delivering them to the destination host.

- Ethernet = is a network access protocol that describes two primary functions:

1) Communication over a data link.

2) Physical transmission of data on the network media.

- Network access protocols are responsible for taking packets from IP and formatting them to be transmitted over media.

Protocol Suites and Industry Standards

- A protocol suite is a set of protocols that work together to provide comprehensive network communication services. (See CH3F3).

- A protocol suit may be specidied by a standards org or developed by a vendor.

- The TCP/IP protocol suite is an open standard, meaning these protocols are free to the public. This allows the protocols to be used by all vendors on their software and hardware.

- Standards in networking ensure that different vendor products can interconnect with products from a separate vendor.

- Some protocols are proprietary which means one company or vendor owns the definitions of the protocol.

- It is also possible for a vendor to develop a protocol and later assist in it becoming a standard.

Development of TCP/IP

- The first packet switching network and predecessor to today's internet was the Advanced Research Projects Agency Network (ARPANET) which was developed in 1969 and connected mainframe computers at 4 locations.

- ARPANET was funded by the US department of Defence for use by universities and research labs.

TCP/IP Protocol Suite

- The TCP/IP suite contains many protocols (See CH3F4):

1) Application layer

- DNS = Domain Name System (or Service) translates domain names into IP addresses.

- BOOTP = Bootstrap Protocol, Enables a disk-less workstation to discover its own IP, the IP of a BOOTP server on the network and a file to be loaded into memory to boot the machine.

- DHCP = Dynamic Host Configuration Protocol, Dynamically assigns IP addresses to client stations at start-up. Allows addresses to be re-used when no longer needed.

- SMTP = Simple Mail Transfer Protocol, Enables clients to send email to a mail server. Enables servers to send email to other servers.

- POP = Post Office Protocol version 3 (POP3), Enables clients to retrieve email from a mail server. Downloads email from the mail server to the desktop.

- IMAP = Internet Message Access Protocol, Enables clients to access email stored on a mail server. Maintains email on the server.

- FTP = File Transfer Protocol, Sets rules that enable a user on one host to access and transfer files to and from another host over the network. A reliable, connection-orientated and acknowledged file delivery protocol.

- TFTP = Trivial File Transfer Protocol, A simple, connectionless FTP. A best-effort, unacknowledged file delivery protocol. Utilizes less overhead than FTP. 

- HTTP = Hypertext Transfer Protocol, Set of rules for exchanging text, graphic images, sound, video and other multimedia files on the world wide web.

2) Transport layer

- UDP = User Datagram Protocol, Enables a process running on one host to send packets to a process running on another host. Does not confirm successful datagram transmission.

- TCP = Transmission Control Protocol, Enables reliable communication between processes running on separate hosts. Reliable, acknowledged transmissions that confirm successful delivery.

3) Internet layer

- IP = Internet Protocol, Receives message segments from the transport layer. Packages messages into packets. Addresses packets for end to end delivery over an internetwork.

- NAT = Network Address Translation, Translates IP addresses from a private network into globally unique public IP addresses.

- ICMP = Internet Control Message Protocol, Provides feedback from a destination host to a source host about errors in packet delivery.

- OSPF = Open Shortest Path First, Link-state routing protocol. Hierarchical design based on areas. Open standard interior routing protocol.

- EIGRP = Enhanced Interior Gateway Routing Protocol, Cisco proprietary routing protocol. Uses composite metric based on bandwidth, delay, load and reliability.

4) Network Access layer

- ARP = Address Resolution Protocol, Provides dynamic address mapping between an IP address and a hardware address.

- PPP = Point-to-point protocol, provides a means of encapsulating packets for transmission over a serial link.

- Ethernet = Defines the rules for wiring and signalling standards of the network access layer.

- Interface drivers = Provides instructions to a machine for the control of a specific interface on a network device.

TCP/IP Communication Process

- Data is encapsulated by the required application protocols (for example a HTTP header if the data is in relation to a web page)

- This is then encapsulated by a transport layer header (for example a TCP header)

- Then this segment is encapsulated by an IP header creating an IP packet.

- Finally a network access layer protocol such encapsulates the IP packet (In the case of Ethernet a header and trailer are added creating a frame).

- The frame is then transmitted to its nearest router where it removes the Ethernet frame and examines the IP packet.

- With this information it then adds its own Ethernet frame and delivers the frame to the router best specified through its analysis of the IP packet.

- This process is then repeated until the data reaches it's destination.

Open Standards

- Open standards encourage interoperability, competition and innovation.

- Also prevents one company monopolizing the market and having an unfair advantage.

Internet Standards

- Standards organisations are generally vendor neutral and not for profit.

- Organizations include:

1) Internet Society (ISOC) = Responsible for promoting open development and evolution of internet use throughout the world.

2) Internet Architecture Board (IAB) = Responsible for the overall management and development of internet standards.

3) Internet Engineering Task Force (IETF) = Develops, updates and maintains TCP/IP technologies.

4) Internet Research Task Force (IRTF) = Focused on long term research related to internet and TCP/IP protocols.

5) Internet Corporation for Assigned Names and Numbers (ICANN) = Based in US, coordinates IP address allocation, the management of domain names and assignment of other information used in TCP/IP protocols.

6) Internet Assigned Numbers Authority (IANA) = Responsible for overseeing and managing IP address allocation, domain name management and protocol identifiers for ICANN.

Electronics and Communications Standard Organisations

- Other standard organizations have the responsibility of developing electronic and communication standards used to deliver IP packets as electronic signals over wired and wireless medium.

1) Institute of Electrical and Electronics Engineers (IEEE) = Dedicated to advancing technological innovation in a variety of fields such as power and healthcare. (See CH3F5 for a list of networking standards).

2) Electronic Industries Allience (EIA) = Standards related to electrical wiring, connectors and the 19-inch racks used to mount equipment.

3) Telecommunications Industry Association (TIA) = Responsible for developing communication standards in a variety of fields including radio, cellular towers, VoIP and satellite. 

4) International Telecommunications Union-Telecommunications Standardization Sector (ITU-T) = One of the largest and oldest standards organisations. Defines standards for video compression. IPTV, broadband and DSL.

Benefits of Using a Layered Model

- The Protocol Model = Closely matches the structure of a protocol suite. The TCP/IP model is a protocol model as it describes the functions at each layer of protocols in the TCP/IP model. it can however be used as a reference model.

- Reference model = Provides consistency with all types of network protocols and services as it describes what needs to be done at each layer but not how it needs to be done. The OSI model is an internetwork reference model but can also be a protocol model.

- For example (see CH3F6).

The OSI Reference Model

- The OSI model provides an extensive list of functions and services at each layer.

- Describes the interaction between each layer and the layer above them.

- The 7 layers are:

1) Physical = The physical layer protocols determine how the bits are transmitted on the physical link. As such are concerned with the mechanical and electrical means to start, stop and maintain the transmission.

2) Data Link = The data link layer describes protocols for selecting best methods for frame transmission between two points based on the media connecting them.

3) Network = The network layer provides the services required to exchange data between two end devices on the network.

4) Transport = Defines the services required to segment, transfer and reassemble data for individual communication between the end devices.

5) Session = Provides services to the presentation layer to organize and manage data exchange.

6) Presentation = Provides common representation of the data transferred between application layers.

7) Application = Contains protocols required for process-to-process communications.

The TCP/IP Model

- Created in the early 1970's, often referred to as the internet model.

- TCP/IP model contains 4 layers:

1) Network Access = Controls the hardware devices and media that make up the network.

2) Internet = Determines the best path through the network.

3) Transport = Supports communication between devices on diverse networks.

4) Application = Represent data to the user as well as encoding and dialog control.

OSI and TCP/IP Comparison

- The protocols that make up the TCP/IP model can be described by the layers of the OSI model.

- In the OSI model the network layer and the application layer of the TCP/IP model are further divided to better describe their functions.

- At the network access layer the TCP/IP model does not describe which protocols to use when transmitting over a physical medium. Only the hand off to the internet layer.

- OSI layers 1 and 2 describe the procedures required in order to access the physical media as well as send the signal over the medium.

- The layer 3 network layer of the OSI model maps directly to the internet layer of the TCP/IP model in which it describes the protocols that address and route data through a network.

- The layer 4 transport layer also maps directly to the Transport later on the TCP/IP model as it describes the general services required to reliably deliver data between two devices.

- The TCP application layer  includes protocols that specify functionality with particular end user applications. The OSI layers 5, 6 and 7 are a reference for application software developers in order to create products for network use.

Message Segmentation

- Theoretically it is possible to send messages in one piece over the network. however in doing his no one else would be able to send messages on the network.

- Additionally if a link were to fail then the message would have to be resent in full instead of a replacement frame.

- In order to prevent this data is broken into smaller pieces, this is called segmentation and has 2 benefits:

1) Smaller segments means that many different conversations can be conducted and interleaved on the network. this is called multiplexing.

2) Can increase the efficiency of network communications, if a message fails only the missing parts need to be retransmitted.

- each segment must be preciously addressed and labelled so that it can be reconstructed into the full message at the end host. 

Protocol Data Units

- As application data is passed down the protocol stack, information is added at each level. this is called encapsulation.

- At any layer the form of this data is called a PDU (protocol data unit).

- As each layer receives the PDU from the layer above it, it encapsulates the PDU with its own data in accordance with the protocols being used or are required.

- Each layer however has a specific name for the PDU:

1) Data = The general term used for the PDU used at the application layer.

2) Segment = The transport layer PDU

3) Packet = Network layer PDU

4) Frame = Data link PDU

5) Bits = Physical layer PDU being transmitted over a medium.

- De-encapsulation = is the reverse process when data is received it moves up the stack to the application, with each header and trailer being removed until only the application data is left.

Network Addresses

- The network and data link layers are responsible for delivering data from a source to destination device.

- Protocols at both layers contain source and destination addresses but for different purposes (see CH3F7).

1) Network Layer addresses = Responsible for delivering the IP packet from source to final destination. This can be on the same network or a remote network.

2) Data Link Layer addresses = Responsible for delivering data from one NIC to another NIC on  the network.

- IP packet contains 2 addresses = The destination and the source.

Data Link Addresses

- data link addresses or Layer 2 Addresses role is to deliver data from one network interface to another on the same network.

- Before an IP packet can be sent over a wired or wireless network. It must first be encapsulated in a data link frame so it can be transmitted over a physical medium.

- As an IP packet moves from host to router, router to router and router to host in is encapsulated in a new data link frame at each step.

- Each data link frame contains the address of the sending NIC and the address of the destination NIC.

Devices on the Same network

- Network layer addresses contain two portions:

1) Network Portion = Identifies the network a device belongs to. All devices on the same network will have the same address.

2) Host Portion = Identifies a specified host. All hosts on the network must have a unique host identifier. 

- The Subnet mask is used to identify the network portion and host potion of these addresses.

- When on the same network the data link layer addresses are MAC addresses on an Ethernet network.

- MAC addresses are physically embedded on the NIC.