Basics Of Computer And Multimedia Networks - MULTIMEDIA

OSI Network Layers

It has long been recognized that network communication is a complex task that involves multiple levels of protocols. A multilayer protocol architecture was thus proposed by the International Organization for Standardization (ISO) in 1984, called Open Systems Interconnection (OSI), documented by ISO Standard 7498. The OSI Reference Model has the following network layers:

  1. Physical Layer. Defines electrical and mechanical properties of the physical interface (e.g., signal level, specifications of the connectors, etc.); also specifies the functions and procedural sequences performed by circuits of the physical interface.
  2. Data Link Layer. Specifies the ways to establish, maintain, and terminate a link, such as transmission and synchronization of data frames, error detection and correction, and access protocol to the Physical layer.
  3. Network layer. Defines the routing of data from one end to the other across the network, such as circuit switching or packet switching. Provides services such as addressing, internetworking, error handling, congestion control, and sequencing of packets.
  4. Transport layer. Provides end - to - end communication between end systems that support end - user applications or services. Supports either connection - oriented or connectionless protocols. Provides error recovery and flow control.
  5. Session layer. Coordinates interaction between user applications on different hosts, manages sessions (connections), such as completion of long file transfers.

Comparison of OSI and TCP / IP protocol architectures and sample protocols

Comparison of OSI and TCP / IP protocol architectures and sample protocols

  1. Presentation layer. Deals with the syntax of transmitted data, such as conversion of different data formats and codes due to different conventions, compression, or encryption.
  2. Application layer. Supports various application programs and protocols, such as FTP, Telnet, HTTP, SNMP, SMTP / MTME, and so on.

TCP / IP Protocols

The OSI protocol architecture, although instrumental in the development of computer networks, did not gain full acceptance, due largely to the competing and more practical TCP / IP set of protocols. TCP / IP protocols were developed before OSI and were funded mostly by the U.S. Department of Defense. They become the de facto standard after their adoption by the Internet,

The above figure compares the OSI and TCP / IP protocol architectures. It can be seen that TCP / IP reduced the total number of layers and basically merged the top three OSI layers into a single application layer. In fact, TCP / IP is even so flexible as to sometimes allow application layer protocols operating directly on IP.

Transport Layer: TCP and UDP. TCP and UDP are two transport layer protocols used in TCP / IP to facilitate host - lo - host (or peer - to - peer) communications.

  1. Transmission Control Protocol (TCP).TheTransmission Control Protocol(TCP) is one of the core protocols of the Internet Protocol Suite. TCP is one of the two original components of the suite, complementing the Internet Protocol (IP), and therefore the entire suite is commonly referred to as TCP / IP. TCP provides reliable, ordered delivery of a stream of octets from a program on one computer to another program on another computer. TCP is the protocol used by major Internet applications such as the World Wide Web, email, remote administration and file transfer. Other applications, which do not require reliable data stream service, may use the User Datagram Protocol (UDP), which provides a datagram service that emphasizes reduced latency over reliability.
  2. TCP is connection - oriented. it provides reliable data transfer between pairs of communicating processes across the network. It handles the sending of application data to the destination process, regardless of datagram or packet size. However, TCP / IP is established for packet - switched networks only. Hence, there are no circuits, and data still have to be packetized. TCP relies on the IP layer for delivering the message to the destination computer specified by its IP address. It provides message packetizing, error detection, retransmission, packet resequencing, and multiplexing.

    Since a process running TCP / IP is required to be able to establish multiple network connections to a remote process, multiplexing is achieved by identifying connections using port numbers. For every TCP connection, both communicating computers allocate a buffer called a window to receive and send data. Flow control is established by only sending data in the window to the destination computer without overflowing its window. The maximum data that can be transmitted at a time is the size of the smaller window of the two computers.

    Each TCP datagram header contains the source and destination ports, sequence number, checksum, window field, acknowledgment number, and other fields.

    1. The source and destination ports are needed for the source process to know where to deliver the message and for the destination process to know where to reply to the message (the address is specified in the IP layer).
    2. As packets travel across the network, they can arrive out of order (by following different paths), be lost, or be duplicated. A sequence number reorders arriving packets and detects whether any are missing. The sequence number is actually the byte count of the first data byte of the packet rather than a serial number for the packet.
    3. The checksum verifies with a high degree of certainty that the packet arrived undamaged, despite channel interference. If the calculated checksum for the received packet does not match the transmitted one, the packet is dropped.
    4. The window field specifies how many bytes the current computer's buffer can receive. This is typically sent with acknowledgment packets.
    5. Acknowledgment (ACK) packets have the ACK number specified - the number of bytes correctly received so far in sequence (corresponding to a sequence number of the first missing packet).

    The source process sends datagrams to the destination process up to the window number and waits for ACKs before sending any more data. The ACK packet will arrive with new window number information to indicate how much more data the destination buffer can receive. If ACK is not received in a small time interval, specified by retransmission timeout (RTO), the packet is resent from the local window buffer. TCP / IP does not specify congestion control mechanisms, yet every TCP / IP implementation should include it.

    Although TCP is reliable, the overhead of retransmission is often viewed as too high for many real - time multimedia applications, such as streaming video. These will typically use UDP.

  3. User Datagram Protocol (UDP). The User Datagram Protocol (UDP) is one of the core members of the Internet protocol suite, the set of network protocols used for the Internet. With UDP, computer applications can send messages, in this case referred to as datagrams, to other hosts on an Internet Protocol (IP) network without requiring prior communications to set up special transmission channels or data paths. The protocol was designed by David P. Reed in 1980 and formally defined in RFC 768.

UDP uses a simple transmission model with a minimum of protocol mechanism. It has no handshaking dialogues, and thus exposes any unreliability of the underlying network protocol to the user's program. As this is normally IP over unreliable media, there is no guarantee of delivery, ordering or duplicate protection. UDP provides checksums for data integrity, and port numbers for addressing different functions at the source and destination of the datagram.

UDP is suitable for purposes where error checking and correction is either not necessary or performed in the application, avoiding the overhead of such processing at the network interface level. Time - sensitive applications often use UDP because dropping packets is preferable to waiting for delayed packets, which may not be an option in a real - time system. If error correction facilities are needed at the network interface level, an application may use the Transmission Control Protocol (TCP) or Stream Control Transmission Protocol (SCTP) which are designed for this purpose.

A number of UDP's attributes make it especially suited for certain applications.

  1. It is transaction-oriented, suitable for simple query-response protocols such as the Domain Name System or the Network Time Protocol .
  2. It provides datagrams, suitable for modeling other protocols such as in IP tunneling or Remote Procedure Call and the Network File System .
  3. It is simple, suitable for bootstrapping or other purposes without a full protocol stack, such as the DHCP and Trivial File Transfer Protocol .
  4. It is stateless, suitable for very large numbers of clients, such as in streaming media applications for example IPTV
  5. The lack of retransmission delays makes it suitable for real - time applications such as Voice over IP , online games , and many protocols built on top of the Real Time Streaming Protocol .
  6. Works well in unidirectional communication, suitable for broadcast information such as in many kinds of service discovery and shared information such as broadcast time or Routing Information Protocol

Network Layer: Internet Protocol (IP). The IP layer provides two basic services: packet addressing and packet fragmentation. Point - to - point message transmission is readily supported within any Local Area Networks (LANs), and in fact, LANs usually support broadcast. However, when a message needs to be sent to a machine on a different LAN, an intermediate device is needed to forward the message. The IP protocol provides for a global addressing of computers across all interconnected networks, where every networked computer (or device) is assigned a globally unique IP address.

For an IP packet to be transmitted across different LANs or Wide Area Networks (WANs), gateways or routers are employed, which use routing tables to direct the messages according to destination IP addresses. A gateway is a computer that usually resides at the edge of the LAN and can send IP packets on both the LAN network interface and the WAN network interface to communicate with other interconnected computers not on the LAN. A router is a device that receives packets and routes them according to their destination address for the same type of network.

The IP layer also has to translate the destination IP address of incoming packets to the appropriate network address. In addition, routing tables identify for each destination IP the next best router IP through which the packet should travel. Since the best route can change depending on node availability, network congestion and other factors, routers have to communicate with each other to determine the best route for groups of IPs. The communication is done using Internet Control Message Protocol (ICMP).

IP is connectionless; it provides no end - to - end flow control. Every packet is treated separately and is not related to past or future packets. Hence, packets can be received out of order and can also be dropped or duplicated.

Packet fragmentation is performed when a packet has to travel over a network that accepts only packets of a smaller size. In that case, IP packets are split into the required smaller size, sent over the network to the next hop, and reassembled and resequenced there.

In its current version, IPv4 (IP version 4), IP addresses are 32 - bit numbers, usually specified using dotted decimal notation (e.g., 128.77.149.63 = 10000000 01001101 10010101 00111111).The 32 - bit addressing in principle allows 232 ≈ 4 billion addresses, which seemed more than adequate. In reality, however, we could be running out of new IP addresses soon (projected in year 2008).

This is not only because of the proliferation of personal computers and wireless devices but also because IP addresses are assigned wastefully. For example, the IP address is of the form (network number, host number). Under many network numbers, the percentage of used host numbers is relatively small, not to mention some inactive hosts that may still occupy their previously assigned addresses.

As a short - term solution to the shortage of IP address availability (due to limitations of service provider or cost), some LANs use proxy servers or Network Address Translation (NAT) devices that proxy servers implement (in addition to content caching and other features). The NAT device separates the LAN from the interconnected network and has only one IP address to handle the communication of all the computers on the LAN. Each computer on a LAN is assigned a local IP address that cannot be accessed from the interconnected network. The NAT device typically maintains a dynamic NAT table that translates communication ports used with its public IP address to the ports and local IP addresses of the communicating computers.

When a local computer sends an IP packet with the local address as the source, it goes through the NAT device, which changes the source IP address to the NAT device IP address that is global. When an IP packet arrives on some communication port to the NAT IP address, the destination address is changed to the local IP address according to the NAT table, and the packet is forwarded to the appropriate computer.

In January 1995, IPv6 (TP version 6) was recommended as the next generation IP (IPng) by the Internet Engineering Task Force (IETF) in its Request for Comments (RFC) 1752, 'The Recommendation for the IP Next Generation Protocol". Among many improvements over IPv4, it adopts 128 - bit addresses, allowing 2128≈ 3.4 x 1038 addresses. This will certainly settle the problem of shortage of IP addresses for a long time (if not forever).


All rights reserved © 2018 Wisdom IT Services India Pvt. Ltd DMCA.com Protection Status

MULTIMEDIA Topics