Acronyms and Glossary
OSI – Open System Interconnection
UDP – User Datagram Protocol is one of the important transport layer protocol used for internet that allows computer applications to send messages.
TCP – Transmission Control Protocol is one of another internet communication protocol
DSL – Digital Subscriber Line is a technology that provides an internet access and transmits digital data over the local telephone line.
UTP – unshielded twisted pair widely used for local area networks
Hub – A hub is a device to connect multiple Ethernet devices
RJ45 – A connector standard used for Ethernet connections and telephone cables
Lindley Hall is a steel and limestone building and has 5 stories. The client wants to connect approximately 1000 computers in the building. Out of these 1000 computers, 900 will be wired and 100 wireless. For this purpose, a consultant has been hired to know about the feasibility of installing a combined network infrastructure i.e. wired and wireless or Ethernet and 802.11.
This report is intended to explain all the terminologies that will be used in developing the network design of the building. It will also explain all network specifications, advantages and disadvantages of using Ethernet and 802.11 independently or combined, necessary equipment, implementation of the design, or any other potential problems that can occur due to any specific type of network.
What is Network for?
Network is a connection of two or more computers by which they can share data and resources. In broad terms, network connects computers, servers and other resources in the entire building. In case of 5 story Lindley Hall building, 1000 computers will be connected together including file servers, printers, and other data processing machines. By means of this network, administration can share a centralized database and each machine may also be enabled with an internet.
In order to design a network for Lindley Hall, first the routine processing at the campus must have to be examined. Who are the users and what applications do they run on the machines? What type of data they are allowed to access? What are their requirements and so on? In case of Lindley Hall, the type of users is students, administrators and staff members. The level of security is different for each so different account will be allocated to each user. Maintenance of the network is highly important as single problem can collapse the entire infrastructure. There are certain resources that are being installed only to manage and maintain other resources.
For determining the feasibility of combined network of Ethernet and 802.11 at Lindley, there is a need to consider basic problems, features, consequences, and implementations of the network design and models. For this purpose, OSI model is used to examine the problems and determine better solutions or protocols that should be used against specific issues.
Today, almost all of the networking models are based on some method of Open Systems Interconnection or OSI standard. This OSI standard was first developed in 1984 by International Organizational Standards organizations. OSI Reference Model consists of seven layers that describe how data travels from one device to another in a network. The lowest in the hierarchy is the physical layer and moving up with data link layer, network layer, transport layer, session layer, presentation layer, and application layer. Each layer processes the data and prepares it for the next layer. The lower layers of OSI model represent technologies like Ethernet whereas the upper layers represent application protocols like UDP and TCP.
7 layers of OSI Model
Application layer being at the top is the final layer that interacts with the operating system or application when the user performs certain network related activity i.e. transferring or reading files. Presentation layer converts the format of data received from the application layer into the one that is understandable by rest of the layers. Session layer carries out communication through receiving device. The transport layer maintains data flow control by integrating data from multiple application sources into single stream. The network layer handles the procedure, data is sent to the recipient. It determines which logical protocols, addressing or routing technique is suitable.
Data Link Layer
The data link layer decides the type of network to be used for the particular infrastructure and thus appropriate physical protocol is allocated to the data. There are two sub layers of data link layer including media access control (MAC) and logical link control (LLC). This layer sets up the connection, divides and converts output data into data frame, manages the acknowledgements received from the recipient, and also analyzes bit patterns, if they are received correctly.
The last layer that is physical layer defines the physical characteristics of network i.e. timing, voltage levels, and connections. The physical layer is dedicated for performing various function but the basic functions of physical layer are as follows:
- Bit representation
- Physical characteristics
- Transmission rate
- Physical topology
- Transmission mode
There are various kinds of hardware technologies that can be chosen for a single type of network architecture or to perform a specific function. It is therefore considered as the most complex layer of the model. However, following devices are commonly being used at organizations and buildings
- Connectors – UTP Connector, coaxial cable connector and fiber optic cable connector.
- Cables – twisted pair, coaxial cable, and fiber optic
- Passive hub – multiport repeater
- Repeaters – repeat signals and thus boost digital transmission
- Simple active hub
Ethernet was first developed for only to connect a printer with the computer but later it expanded and today it is the base of every modern network. Ethernet has so far evolved various other technologies; however, the basic idea is still the same i.e. to connect devices to each other with a single cable. By means of this single cable Ethernet, new devices can be added to the network without any design modification.
Ethernet is also called as local area technology since it is used to connect devices that are in close proximity with each other. Usually Ethernet networks are deployed in homes or small buildings. It is not as such suitable for large distance areas i.e. cities or two buildings. Since, there is one cable that connects all devices so it is not possible to plot Ethernet network up to large areas.
In an Ethernet network devices are attached with a single cable; therefore, a signal sent by one device passes through all other devices connected with the cable. In order to identify a certain device, it is allocated with a unique Internet Protocol or IP address. Thus, when computer A sends a print command, signal is carried along with an address and reaches the printer.
However, whether it is wired or wireless connection, there is one thing common between both types of networks. Both networks make use of hardware and software items to transfer information from one device to another. However, information sent over other device does not reach randomly but there is a unique address for each device. Internet Protocol which is based on network software is not the only unique address but there is also a unique hardware address. This address is usually stored in the network interface card. There is 12 bit hexadecimal number carrying 48 bits. The first six bits show ID number of the network adapter and the second half of the adapter show the serial number adapted by the manufacturer.
In an Ethernet cable, data is sent over the devices as electricity and as like the properties of electrical currents, the signal gets weaker as it travels further. However, the quality of Ethernet cables has improved a lot with time and so the range of Ethernet networks has also been improved. In addition, repeaters are also being used at Ethernet networks to boost up the signal strength by repeating it .
Different types of coaxial cables are being used for an Ethernet network that supports electric currents. In a coaxial cable, there is a central wire which is wrapped by plastic dielectric, then metallic shield and finally an outer jacket. The central wire is usually made of copper and is braided. The electrical and magnetic fields are being created when current passes through the central wire. These electrical and magnetic fields block any outer interference. Dielectric is coated because the metal shield further prevents outer electric or magnetic interference. The shield is either in form of metal foil or braided wire mesh. The later is more common because adds flexibility and improved insulating properties to the cable.
Standard RJ45 is a UTP connector that is used to connect Ethernet physical layer device. It is usually being used with the CAT5 cable i.e. for Ethernet networks. Devices can be connected by bus, star, or point to point network topology. Network topology is the specific arrangement of network devices that is followed over the entire network.
Advantages of Ethernet Network
- Ethernet network is cheaper than 802.11 network
- It is easier to maintain and administer Ethernet network
- Ethernet cables are noise resistant so there are least external disturbances
- Communication over Ethernet cable is reliable and secure
- Easier to connect new devices to Ethernet
Disadvantages of Ethernet Network
- There is limited bandwidth on a single LAN since all devices share it. So, only limited number of devices can be added.
- Ethernet has a limited length.
- Signals sent through one device can collide with other stations.
- Variable delays occur due to packet switching
- There is no option of priority mechanism
- Installation of Ethernet network is difficult
Bus and Star Topology Advantages and Disadvantages
Based on the network topology, there are different advantages and disadvantages associated with each type of design. In bus topology, Ethernet cable is deployed in shape of bus. Data can be transmitted in only one direction at a time. All devices on the bus are connected on local area network and the central cable is considered as the backbone of the network. If the cable gets damaged from anywhere, the whole network gets collapsed. However, it is easiest to install and has initial costs than other topologies. It can be deployed in any smaller or bigger networks .
In comparison to the bus topology, star topology is bit expensive as it requires more cabling. In this type of topology, the devices are connected with the central computer that acts as hub. Star topology is more reliable than other topologies and upon damage, the rest of the network keeps on operating. However, if the central hub fails to work, then the whole network gets collapse. Another bigger benefit of this topology is that, any device can connect or communicate with the other device on the network. The packets do not need to pass through unintended recipients but directly reaches to the intended machine. Thus, it further secures the system and also makes it more efficient.
In case of Lindley Hall, star topology is a better choice as compared to bus topology. It is due to reason that there are 1000 computers in a campus that needs to be on the network. If they are connected via bus topology, then the processing and speed will get extremely slow. Therefore, in order to speed up the network, it is recommended to deploy star topology. Though, initial expenses will be higher but later, it will speed up operations.
Wireless network or 802.11
Wireless technology (802.11) has become most popular nowadays mainly because of its large area coverage without use of wires. It is also called as wireless local area network protocol or WLAN. Wireless network make use of radio waves for communication between the devices. The wireless adapter translates data in form of radio signal i.e. 1s and 0s and transmits it by means of an antenna. On the other end wireless router receives signal and then decodes it. This decoded information is then sent over the internet using wired Ethernet connection. The same process also works in reverse. Each 802.11 standard transmits signal at different frequencies .
Initially, WLAN was incapable of supporting bandwidth above 2Mbps but now with the 802.11n standard maximum speed is supported. During this time, various standards were introduced including 802.11a, 802.11b, 802.11g, and 802.11n. Users configure and enable wired equivalent privacy (WEP) at the access point to prevent network from unwanted intrusions. However, this technology was found to be less efficient in securing the network and so other technologies including 802.1X protocol and Wi-Fi Protected Access (WPA) were introduced to secure the network infrastructure.
Media Access Control or MAC address filtering is another technique to secure the network. It uses computer’s physical hardware to authenticate users. The unique MAC address of each device is used for filtration. Administrator lists down the addresses of the machines that are allowed to access the network .
Wireless Network Requirements
In order to build a wireless network at any building or home, following units are required:
- Wireless router – receive and broadcast radio signals to the network devices and users.
- Port to connect to DSL modem or cable
- Ethernet hub
- Wireless access point
In order to receive signals through the wireless network, the devices must be equipped with the network card or adapter that can catch radio signals.
Advantages of Wi-Fi (802.11) Network
- Temporary network can be created for small period of time and later it can be moved to other place.
- No mess of wiring and so less maintenance costs in incurred.
- Provides portability to the users. Students or staff can connect internet to any device i.e. laptop or mobile while being anywhere in the campus.
- Multiple users can communicate with the single machine at a time
- Single network can also communicate with number of devices at a time .
Disadvantages of Wi-Fi (802.11) Network
- Less secure than Ethernet as other users can also access the network if security protocols are not configured.
- Speed is slower than the wired or Ethernet network in most of 802.11 standards.
- Initial installation costs of Wi-Fi networks are usually higher than the Ethernet network. However, with the popularity of 802.11 standards, the prices of Wi-Fi network devices are dropping.
- Less reliable than Ethernet network .
Combination of Ethernet and Wireless Network
In case of Lindley Hall where there is a need to connect 900 computers via an Ethernet and 100 computers to be connected through Wireless or 802.11 network would have a combination of both networks. It is possible to design a combination of both these networks at Lindley Hall. However, some problems may occur in this type of network design.
In order to deploy an Ethernet network at Lindley Hall, 900 computers will be connected via an Ethernet cable. There are five stories in a building so, it can be assumed that there are 200 computers at each floor. Ethernet cable can be connected from one floor to other floor as well so five floors will be connected via an Ethernet network. The majority computer users are students so four floors will be reserved for students and 1 floor for the staff members and administrators.
Any device that wants to connect to the wireless network needs to have wireless adapter. Nowadays, most laptop machines and computers have network adapters already built in them. So, 100 computers at Lindley Hall that needs to access an internet via wireless network must have wireless adapter. Wireless subscription will only be provided to the staff members and administrators. Wireless router will be connected to the DSL modem and then configured with security protocols, firewall, and other technologies. MAC address range and filtration will also be configured so that no outside users can access the campus network.
Common wireless routers have 4 to 8 slots available for wired Ethernet connection. But, there needs to have 900 wired connections so for this purpose, N-port Ethernet switches or hub can be used to expand the number of wired connections. Network switch is the device that is used to connect number of wired devices to the network. Network switch is also called hub. This type of hybrid models is only possible to develop if there is enough number of switches. In this case, the cost would increase due to requirement of large number of switches.
Applegate, Mark. Disadvantages of Ethernet Cabling. n.d. http://www.ehow.com/info_8554691_disadvantages-ethernet-cabling.html.
Brain, Marshall, Tracy Wilson and Bernadette Johnson. How WiFi Works . n.d. http://computer.howstuffworks.com/wireless-network.htm.
Pidgeon, Nick. How Ethernet Works. n.d. http://computer.howstuffworks.com/ethernet5.htm.
Roberts, Steve. "Understanding Ethernet Communication - Topologies." 2011.
Sheldon, Ryan. Introduction to Wi-Fi 802.11b and Ethernet Communications. 2008. https://www.controlanything.com/Relay/Device/A0005.
Smith, Matt. Wi-Fi Vs. Ethernet: Has Wireless Killed Wired? 18 Jan 2013. http://www.digitaltrends.com/computing/wi-fi-vs-ethernet-has-wireless-killed-wired/.