TCP/IP protocol history reaches the 1960s when, from the initiative of the US Department of Defence, a study on a universal computer network that would allow operation also in case of a failure of part of its physical infrastructure. The project has been improved for many years up to its current form. With an increase in popularity of the internet, TCP/IP has become the main communication protocol.
TCP/IP is an open protocol not limited by copyrights. This means that anyone can see how it functions and implement it in a software. It is both the reason for its success and the main source of security risk.
The design and basic working principle of the protocol are relatively complex. In simple terms, the protocol can be used to divide the transmitted data into packets and to identify those packets for easy verification and joining on the receiving side. It is critical, since the packets travel individually following the most optimum path available. Each packet must include information about the sender, the receiver, the type of transmitted data and TTL to limit the lifetime of packet when attempting to contact an address.
In simple terms, that is how the data are transmitted in the computer networks. In practice, the TCP/IP protocol is so complex that it has four layers used for different tasks.
A short specification is shown below.
Physical layer - network access
The main layer receiving data from the next internet layer. It transmits data as a string of bits. In practice, the physical layer is a network adaptor or a modem with drivers installed in the operating system.
It transport data packets from the server to the target device and vice versa. Key procedures related to establishing the connection between the devices are carried out in this layer. It does not guarantee correct data transfer.
It provides communication between user applications. It controls the information flow and provides uninterrupted transfer of packets in a specific order. The data are divided into packets and identified with the sender, receiver and data type information. The receiving device sends a confirmation to the server after all packets are received. Otherwise, the server will attempt to send the data package again. This layer is used by internet apps, e.g. internet browser that cannot directly communicate with the network.
The highest layer that allows applications to access the services of the other layers. The layer includes a suite of ready-to-use protocols used by application for network communication including HTTP, TELNET, POP3 and SMTP.
TCP/IP is a suite of different protocols intended for use in different layers.
Correct functioning of the entire system also requires correct IP addressing. Each device connected to the internet must have a unique IP address. It applies to all the devices with direct access to the internet, e.g. modems or network adapters. IP addresses are divided into 5 classes and assigned depending on the type and application. IP address pools are delegated by nominated organisations to ISPs so that they can assign them to their customers. The organisation maintain order and hierarchy of addressing to prevent the addresses from being used more than once. In Poland this organisation is the NASK. Some IP addresses has been reserved, which means that the address cannot be found in the internet. For example, addresses from a 192.168.x.x. pool, where x is any one, two or three-digit number are used in local area networks only.
We should also mention the DNS (Domain Name System) servers that are used as translators for domain names. With DNS servers, instead of a 32-bit IP address, we can enter a website address www.delta.poznan.pl, which is much easier than having to memorize a long string of numbers. The rest is handled by the nearest DNS server with a list of website addresses and their corresponding IP addresses. Since the DNS structure is decentralized, a real-time communication and exchange of information is required between DNS servers and newly created domains.