Demystifying the TCPIP Protocols

Demystifying the TCP/IP Protocols

The TCP/IP suite is the global standard for Internet communications. TCP/IP isn’t a single protocol, but a suite containing dozens of protocols. The name refers to two of the most important protocols in the suite, the Transmission Control Protocol (TCP) and the Internet Protocol (IP). IP provides addresses that make TCP/IP communications routable. This means that all messages transmitted over a TCP/IP network are marked with the IP address of the destination network and station.

TCP is one of the main transport protocols in the suite (the other being UDP). TCP ensures that data arrives at its destination without errors or missing packets.

Defining important protocols

The TCP/IP suite contains dozens of protocols. Because network services and devices use some of them more than others, you are more likely to encounter them whenever you read about networking. Table 2-1 lists some of the more important protocols in the TCP/IP suite, but it’s by no means complete.

Table 2-1: Common Protocols in the TCP/IP Suite

Protocol Name	Function
ARP	Address Resolution Protocol. This protocol matches hardware (MAC) addresses to IP addresses. This protocol is targeted in some of the more technical attacks on Wi-Fi.
DHCP	Dynamic Host Configuration Protocol. Remote configuration of network addresses on client machines (more detailed description later in this chapter).
DNS	Domain Name System. References host names to IP addresses.
FTP	File Transfer Protocol. Describes transferring files between nodes.
HTTP	Hypertext Transfer Protocol. Used to deliver and read Web pages.
ICMP	Internet Control Message Protocol. Delivers routing control messages.
UDP	User Datagram Protocol. Connects applications between network hosts.

Understanding packet switching and encapsulation

All TCP/IP networks, even wireless ones, use a routing technology called packet switching. On a packet-switched network, computers divide data into smaller, individually addressed packets. Because each of these data packets contains a destination address, they can each follow different paths to reach their destination. Contrast this with a circuit-switched network, such as the traditional telephone system, where communications require a dedicated point-to-point connection (see Figure 2-5).

Figure 2-5: Packet switching versus circuit switching

Because users can share bandwidth, rather than each user requiring its own dedicated circuit, a packet-switched network uses network bandwidth more efficiently. IP packets don’t have to arrive at their destination in the correct order. Packets contain sequencing information in their address headers. When they arrive at their destination, the receiving host reassembles the packets into the original message.

As packets move through the network, each successive layer of the TCP/IP model adds its own header information. The network layers encapsulate the data within these successive headers (see Figure 2-6). The headers contain networking instructions and information, including from and to addresses. Encapsulation is analogous to placing the data in an envelope and sending it to its destination. Depending on the destination address and intended program, your computer may encapsulate data within multiple envelopes.

Figure 2-6: Encapsulation within headers

As the encapsulated packet travels to its destination, network layers and services only read the specific header information that instructs them what to do with the packet. As a packet arrives at its destination, each of the headers is stripped away until the data arrives at the destination program. Demultiplexing is the process of encapsulation in reverse.