Tuesday, January 29, 2008

Data Communications

There are two basic types of data communications: circuit-switched data and packet-switched data. Circuit-switched data provides for continuous data signals while packet-switched data allows for rapid delivery of very short data messages.

Circuit-switched Data
Circuit-switched data is a data communication method that maintains a dedicated communications path between two communication devices regardless of the amount of data that is sent between the devices. This gives to communications equipment the exclusive use of the circuit that connects them, even when the circuit is momentarily idle.

To establish a circuit-switched data connection, the address is sent first and a connection (possibly a virtual non-physical connection) path is established. After this path is setup, data is continually transferred using this path until the path is disconnected by request from the sender or receiver of data.

Figure below shows the basic operation that uses circuit-switched data. In this example, a laptop computer is sending a file to a company’s computer that is connected to the public switched telephone network (PSTN). The laptop computer data communication software requests the destination phone number from the user to connect to the remote computer. This telephone number (the address) is used connect a path through the PSTN switches until the call reaches the destination computer. The dialed number is first connected through local switch #1, port number 4236. This port number is assigned to a memory location in the switch that routes the data connection through a high-speed line, time slot 6 to an IXC switch. The IXC switch then assigns a memory location in its switch to a high-speed line, time slot 3 that connects to local switch #2. Local switch #2 assigns a memory location in its switch to port number 1249. This port connects to the remote computer. Once this path through the network is setup, it remains constant throughout the data communications session regardless of how much data is transferred between the laptop computer and the company’s computer.


Circuit-switched Data


Packet-switched Data
Packet data service provides data transfer in the form of short packets of information. The public telephone network was designed primarily to offer voice services. Shortly after the telephone network was introduced, circuit-switched (continuous) data services were offered. The operation requirements for circuit-switched and packet-switched data services are very different. Circuit-switched data has substantial time and is inefficient for serving sensing control and applications that require small amounts of information. Initially the standard telephone system had to be enhanced (functionally divided) to offer packet data service. However, with the digitization of communications systems, telephone systems operate more like packet data systems.

Typical applications for packet data service include Internet browsing, wireless email, train control system, route guidance, credit card processing and many other applications that benefit from the transmission of data in bursts when communicating.

Packet data systems provide effective use of the resources. Packet data systems only use network equipment resources when there is information to transfer. This provides the advantage of charging only for the amount of information used and increased system efficiency.

A packet is a group of digital bits that is transported and switched through a network of packet switches (often called routers) to their destination. The structure of these packets (digital bit sequence) is arranged in a specific format to allow the determination of the destination address for each packet in addition to the data that is being transported. Optionally, the packet structure may include other information such as the packet originator and error protection bits.

Transmitting data through a packet network involves dividing data files into small packets (typically under 100 bytes of information). A packet data system divides large quantities of data into small packets for transmission through a switching network that uses the addresses of the packets to dynamically route these packets through a switching network to their ultimate destination. When a data block is divided, the packets are given sequence numbers so that a packet assembler/disassembler (PAD) device can recombine the packets to the original data block after they have been transmitted through the network.

Figure below shows the basic operation that uses packet-switched data. In this example, a laptop computer is sending a file to a company’s remote computer that is connected to a packet data network. The laptop computer data communication software requests the destination address for the packets for the user to connect to the remote computer (202.196.22.45). In this example, the source computer divides the data file into three parts and adds the packet address to each of the 3 data packets. The packets are sent through routers in the packet network that independently determine the best path at the time that will help the packet reach its destination (smart switches). This diagram shows the three packets take 3 different routes to reach their destination. When the 3 packets reach their destination, the remote computer reassembles the data packets into the original data file.


Packet-switched Data


Public Data Networks (Internet)
Public data networks interconnect data communication devices (e.g. computers) with each other through a network that is accessible by many users (the pubic). To allow many different users to communicate with each other, standard communication messages and processes are used. The Internet is an example of a public data network (there are other public data networks) that uses standard Internet protocol (IP) to allow anyone to transfer data from point to point by using data packets. Each transmitted packet in the Internet finds its way through the network switching through nodes (computers). Each node in the Internet forwards received packets to another location (another node) that is closer to its destination. Each node contains routing tables that provide packet-forwarding information.

Each network in the Internet can have different transmission formats (e.g. different packet sizes, high-speed or low-speed data) but the all agree on how to receive and distribute IP packets. Internet service providers (ISPs) connect users (e.g. computers) to the Internet. ISPs are interconnected to each other through network service providers (NSP). NSPs are relatively large networks that may cross international boundaries. NSPs can connect to each other through network access points (NAPs). Because there are a limited number of NAPs, there are also private network access points (PNAPs). PNAPs are setup by the NSPs to relieve the congestion on the NAPs.

Figure below shows the Internet. This diagram shows that the Internet is composed of users (end points), Internet service providers (ISPs), network service providers (NSPs), and network access points (NAPs). Computers are connected to the Internet via an ISP. The ISP receives data from the computer, reformats it (if necessary), and forwards it to the destination computer in its network. If necessary, it may be routed to an NSP, which will route the data packets to their destination ISP to an NAP that will allow the packet to reach its destination. Eventually, packets reach their destination ISP that forwards the packets to the user.

No comments:

Post a Comment