Analysis Of IP Protocol Communications Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The Transmission Control Protocol (TCP), makes up for IP's deficiencies by providing reliable, stream-oriented connections that hide most of IP's shortcomings. The protocol suite gets its name because most TCP/IP protocols are based on TCP, which is in turn based on IP. TCP and IP are the twin pillars of TCP/IP.

TCP adds a great deal of functionality to the IP service it is layered over:

  • Streams. TCP data is organized as a stream of bytes, much like a file. The datagram nature of the network is concealed. A mechanism (the Urgent Pointer) exists to let out-of-band data be specially flagged.
  • Reliable delivery. Sequence numbers are used to coordinate which data has been transmitted and received. TCP will arrange for retransmission if it determines that data has been lost.
  • Network adaptation. TCP will dynamically learn the delay characteristics of a network and adjust its operation to maximize throughput without overloading the network.
  • Flow control. TCP manages data buffers, and coordinates traffic so its buffers will never overflow. Fast senders will be stopped periodically to keep up with slower receivers.

The most fundamental concept within the TCP/IP is the simple fact that everything that connects to the Internet gets an IP address. The Internet Protocol ensures that all addresses on a given network are related, an essential aspect of how the Internet operates. For example, imagine two network devices with IP addresses of and Of the four octets of their 32-bit addresses, the first three are identical. From that, we can tell that these two devices are on the same subnet. This structure of IP addresses is a critical point, because it tells us to which individual network among the many hundreds of thousands that form the Internet.

Internet Routers

The fundamental building block of the “network of networks” now known as the Internet is a device called an IP router. Essentially, a router connects two or more individual networks so that they can exchange data packets using the Internet Protocol. The basic concept of IP routing is depicted in Figure below. Computers or other endpoint devices connected to one of the individual networks communicate with other endpoints by transmitting and receiving packets of data formatted according to the Internet Protocol. Each packet of data contains the IP addresses of both its source and destination, so the packet can find its way independently through the connected networks to reach its destination. This is a tremendously powerful concept with great benefits for radio-based networks, with their unpredictable connectivity.

The packets are routed from source to destination using IP routing.

Wired Networks

Figure below shows the simple example of a single physical Ethernet network with five computers and a router.

In order for everything to work properly, there are three pieces of information that absolutely must be configured into each computer connected to this network:

  • Its own IP address
  • A number known as a network mask
  • The IP address of the “default router” that connects this network to other networks.

Every IP packet has a source address and a destination address that allows each packet of information to be routed individually from the source to the destination. For any given type of physical network (such as Ethernet, ATM, or X.25), there is some form of address that is usually very different from an IP address. Figure above shows a small Ethernet as an example, so every attached computer has a 48-bit Ethernet MAC address. MAC addresses are quite lengthy, so for simplicity, they are shown in Figure above as simply MAC-A through MAC-F.

The local addressing mechanism is an attribute of each type of network, so the form and specifics are different across different network technologies. Therefore, there needs to be a way of correlating a computer's IP address to its local network address. Each type of network has a different way of doing this. In the case of Ethernet, each computer keeps an internal table that lists both the MAC address and the associated IP address as shown below.

Table 1: IP Address and associated MAC Address


IP Address

Local Network Address

Computer 2


Computer 3


Computer 4


Computer 5




With that information, Computer 1 can communicate directly with any other computer on its local Ethernet network. However, it must communicate with computers on other networks only by sending packets through the router. This is where the network mask and default router configurations come into play. The network mask is a number that helps network devices determine whether a destination IP address is on the local network, or whether it is only reachable via a router. If the latter, then the device must know the address of its default router. With these pieces of information, network devices can properly send their traffic either directly to other local devices, or to the appropriate router for remote delivery.


A single MANET consists of radios that can exchange data over a certain geographic area. They could be relatively short-range radios, or BLOS (beyond line-of-sight) radios with ranges of hundreds of miles. The underlying radio network must support an ad hoc capability in which radios can change in relation to each other, and can enter or leave the radio network. This is, in fact, a form of MANET that is operating at the radio network layer. The focus of Radio Aware Routing is to provide a mechanism for routers to connect and exchange information over radio paths and be able to obtain information about the radio links. Such detailed information about the links between radios will allow the IP routing to select the optimal paths, and enable more rapid convergence of IP routing due to any changes in the underlying radio network. The addressing schemes used in different radio networks will vary, just as they do across different wired networking technologies such as X.25, ATM, and Ethernet. For example, a radio with a broadcast (omnidirectional) antenna may simply use an existing addressing method, such as Ethernet MAC addresses. Other types of radio networks may use something as simple as an 8-bit identifier for each radio terminal. One can also layer addressing schemes on top of each other, which already happens with most networks that transport IP traffic.

If for example, each radio gets a unique 8-bit address, you could have up to 255 radios on a single network. Let's assume that the radios are omnidirectional and can therefore broadcast information to all other radios in range. A simple representation of this notional radio network is shown in Figure below.

In this example, every radio has an individual address, numbered 1 through 6. The diagram also shows that there must be some type of physical interface to the computers (or other endpoint devices) that will use the radios to exchange data. If a new radio comes into range of this group, it must be able to join the network and begin exchanging data with the other radios. Having this occur without manual intervention or reconfiguration of the radio.

A Mobile Ad-hoc Network (MANET) is a temporary wireless network composed of mobile nodes without an infrastructure. A MANET may be suitable for networks within airports, meeting rooms, and open spaces due to both economical and technological feasibilities. Because the MANET is based upon IP protocol suites, a node in the MANET cannot take part in unicast communications until it is configured with a free IP address.

Although it is simple to set IP addresses of mobile nodes in a small scale MANET, it becomes desirable for the procedure to be automatic for a large scale open MANET where mobile nodes are free to join and leave, which has motivated research efforts into the study of autoconfiguration in MANETs.

Another problem associated with IP address assignment of a mobile node is that the IP address may change during its session in the MANET. IP address change is not a serious problem in hardwired networks because the IP address of a host is either statically configured or dynamically allocated by a DHCP server. It usually does not change its IP address during a session unless it reboots. However, because the nodes in the MANET are free to move arbitrarily, IP address change happens more frequently when applied with autoconfiguration, global connectivity, and hierarchical addressing schemes. There are several scenarios in which a mobile node will change its IP address:

  • Merger of two partitions of a network: If some mobile nodes in the MANET move out of the transmission range of the other nodes.
  • Merger of two independent MANETs: The second scenario is that two independently configured MANETs merge. Because these two networks are autoconfigured separately, there may be some duplicate addresses in both networks.
  • Merger of a MANET with a LAN: The third scenario is that a MANET merges with a LAN that has an “ad-hoc” mode Access Point (AP). The mobile nodes that are within transmission range of the AP of the LAN may want to use the configuration information broadcast by the AP to configure itself and function as a relay node. As a result, the MANET becomes connected to the Internet. Furthermore, if the MANET and the LAN use the same private address range, there may be duplicate addresses in both the MANET and LAN.
  • A MANET with a hierarchical addressing scheme: In the network where a hierarchical addressing scheme is deployed, the mobile nodes are divided into different clusters, each of which has a unique subnet address. When a mobile node moves from one cluster to another cluster, it will change its address to one with the corresponding subnet address.








Comparison of TCP Protocol in Wired and MANET Page 7