Study On Computer Network Technologies Computer Science 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.

Star Network refers to the network each node device through a network focused on devices such as the hub or switch HUB Switch connected together, each node star-shaped distribution network connection. This topology is mainly used in IEEE 802.2, IEEE 802.3 Ethernet standard. 

Basic features: 

Easy to implement, but the installation and maintenance workload, cost more: it is generally used in the transmission medium using common twisted pair or coaxial cable. But each site must be focused on the central network device connected directly and requires a lot of cable, and installation and maintenance workload has increased sharply. 

Node extension, mobile convenience: node extensions only concentrated from the hub or switch and other devices to pull a cable, but just need to move a node corresponding to the new node to node equipment. 

Fault diagnosis and isolation easy: a node failure will not affect other nodes connection, can take any failed nodes removed; heavier burden on the central node, easily become a bottleneck; capacity of each site, the lower the distribution: the central node in the event of failure, the entire network was affected.

Label 1.2 Ring Network topology

Ring topology is a network topology where each node is connected to two other nodes, forming a single route for the transmission line directly to each node. The data will go through from each node to every node in turn, each node routes traveled will receive each data packet.

Because the ring topology provides only one path between two nodes, the topology is likely to be disrupted in the event of failure of a link. A node failure or cable damage occurs causing all nodes to separate from the ring topology. FDDI networks can overcome the shortcomings of the data transmission clockwise and counter clockwise to form a ring: in the event of failure data to be sent back to complete the ring before he arrived at the end of the cable, it acts conducted during each of the nodes such as "ring C". 802.5 networking protocol - also known as the IBM token ring network - also avoid these disadvantages: they use a star topology at the physical layer access to the various stations and units to mimic the ring at the data link layer.

Many ring networks to add a "counter rotating ring" to form a redundant topology.


This does not require a computer network server

Networks can be built wider use token ring


A computer malfunction caused the problem in the network

Not easy to repair if damage occurs

MAU expensive type of network card from the card type of Ethernet network

Slower than Ethernet

Label 1.3 Bus Network topology

Bus topology, also known as a linear bus, the network topology is the simplest. It consists of a cable, known as segments or spinal cord, which connects all the computers in the network in a row. Three concepts should be taken into account so that computers can communicate on the bus network. The data in the network is in the form of electronic signals, it is sent to all computers in the network. This information will only be accepted by the computer that has an address that matches the address that is encoded by the original signal. Only one computer can only send signals or data within a period of time.

Question 1

What are three (3) basic modes used in MAC to control access to medium by devices? Describe one of the modes.

Answer 1.B

The MAC (Medium access control) is a sub-layer to closely associate with the physical layer. The three basic modes use in the MAC there are Reservation, Contention and Round Robin.

Reservation is reserve a station to send the data slots for future expansion, or even indefinitely.

Question 1

Explain CSMA/CD with the help of a diagram.

Answer 1.C

Label 1.4 Simplified Algorithm of CSMA/CD

Carrier Sense Multiple Access / Collision Detection (Carrier Sense Multiple Access with Collision Detection, CSMA / CD) 

This program requires the device to send the frame at the same time listens to the channel to determine whether a conflict if the process of sending data conflict is detected, then the following conflict management operations: 

Send a special block and stopped sending data information: Special block a few bytes of information are all one continuous signal, a move intended to strengthen the conflict in order to allow other devices to detect the conflict as quickly as possible. At a fixed time (the beginning is 1 contention period times) to wait a random time, once again sent.  If you still crash, the use of truncated binary exponential algorithm to avoid sending back. Stop within the previous ten second "fixed time" and then double time to send a random ten times to stop before a "fixed time" and then sent randomly. Even after 16 times trying to give up the transmission failure. This program applies to Ethernet (DIX Ethernet V2) standard, IEEE 802.3 standard.

Question 2

IEEE 802.3 Ethernet standard allows different types of cables to be used with a data transfer rate of 10 Mbps. Explain the Ethernet implementations which use coaxial cable, UTP cable. Use a proper diagram with appropriate topology.

Answer 2.A

Ethernet is a data communications framework consisting of a series of standard way of wiring and signal generated and sent across the cable. Ethernet standard is to maintain the Electrical and Electronics Engineers IEEE 802.3 working group to develop a standard, Ethernet-based local area network. Use an Ethernet cable, including those with copper core and those with fiber core, for faster data transfer rate.

Label 2.1 is Star topology in UTP.

Category 5e UTP cable

10 base T Ethernet uses UTP (5e category) unshielded twisted pair (UTP) cable, which is commonly used and relatively inexpensive cables to other network solutions, such as 100 Base T Ethernet network Optical fiber network. Category 5 cable consists of copper core encapsulated in an insulating material made ​​of PVC. Two-core copper wire, insulation and PVC packaging, is twisted around each other along the length of the cable. What is the meaning of the term "twisted pair" of the twisted pair Twisted pair of insulated wire, then covered with insulation jacket, and then from polyvinyl chloride to make up for the CAT 5 cable. Winding wires around each other along the cable to protect data from electromagnetic interference may come from other nearby cables or electrical equipment. Maximum length of a segment of the CAT 5 UTP cable is 328 feet (100 meters). After this, the length of the signal along the cable drop. 10 base T network data transfer speed of 100 megabits per second. UTP is using for Star topology all wiring is done from a central point (the server or hub) and normally STP or UTP are four wires. The star topology has the greatest cable lengths of any topology (and thus uses the most amount of cable). The advantage for the star topology is easy to add new workstations and easy centralized control the network or monitoring the hub. The disadvantages is if hub failure cripples all workstations connected to that hub and hubs are more expensive than thin-Ethernet. Twisted pair of quality may vary from very high-speed cable telephone grade wire. The cable has four pairs of wire inside the jacket. Each pair is twisted a different number of twists per inch to help eliminate interference with adjacent pairs and other electrical equipment.  EIA / TIA (Electronic Industries Association / Telecommunications Industry Association) has established standards of UTP and scoring, six lines (the other categories are emerging.)

Coaxial cable can be used in long distance connection example the building to building or bus network topology.

Label 2.1.1 is bus network topology

Coaxial cable 10BASE 2

Early implementation of Ethernet, called Ethernet 10BASE 2, often used coaxial cable network equipment. RG58 coaxial cable, because of the need to use other types of coaxial cable may have different electrical characteristics, may not work in the data transmission environment. RG58 coaxial cable typically includes a solid copper core, with 20 AWG copper wire. This core is covered with solid polyethylene shield, wrapped in aluminum foil and polyethylene insulation along the length of the cable. The cable is then covered with a flame-retardant PVC jacket. Maximum length of coaxial cable for a 10BASE 2 is 607 feet (185 meters). Coaxial - two conductors separated by insulating layers, such as TV 75 ohm cables. Maximum lengths is 185-500 meters. Thin - thin cable interfaces with the British Navy (BNC) at the end of the year. Thin is part of the RG - 58 series cable *. Maximum cable length is 185 meters. Transfer rate of 10Mbps. thin cable should have 50 ohm impedance, the termination impedance of 50 ohms. T or barrel connector has no impedance.

Thicket - inch semi-rigid cable. The maximum cable length is 500 meters. Transfer rate of 10Mbps. Expensive and not commonly used. (The RG - 11 or RG - 8). A vampire tap or piercing tap water for a transceiver connected to the computer connected to the cable. 100 connections may be made. The computer has an attachment unit interface (AUI connector) in its network card, is a 15-pin DB, 15 connector. Computer is connected to the transceiver AUI cable from the card using the DROP on its network cable.

Coaxial cable type:

The RG - 58 / ü - 50 ohms, with a solid copper core.

The RG - 58 / ü * - 50 ohms, and the core wire.

The RG - 58 of the C / ü * - Military version of RG - 58 / United States

The RG - 59 - 75 ohms, broadband transmission, such as cable TV.

The RG - 62 - 93 ohm, mainly for the Arc Net.

The RG - 6 - for satellite cable (if you want to run the satellite cable!).

* Only these are part of IEEE standard for Ethernet networks.

Label 2.1.2 is Coaxial cable 10BASE 2

Question 2

Explain two different implementations of Fast Ethernet

Answer 2.B

Fast Ethernet is use CSMA/CD in star wire bus topology to run UTP data or fiber optical cable. It also similar with 10baseASE-T cable are attach to a hub. It provides compatibility with existing 10BASE-T system and it also upgrade from 10BASE-T. Fast Ethernet is referrer to 100 BASE-X and the X is a placeholder for the FX and TX variants.

The 100 media type designation is refers to the transmission speed of 100Mbit/s.

The "BASE" is refers to the baseband signaling, which means that only Ethernet signals are carry on the medium. The TX FX and T4 are referring to the physical medium that carries the signal. The fast Ethernet adapter can be logically divide into MAC which deals with the higher level of medium and PHY (Physical Layer Interface). The MAC can be link to the PHY by the 4 bit 25 MHZ synchronous with the parallel interface as a MII or 2 bit 50 MHZ variant (RMII).

The 100 BASE-T is several for the fast Ethernet for twisted pair cables, this will including 100BASE-TX (100 Mbit/s over two pair Cat5) 100BASE-T4 (100 Mbit/s over 4 pair Cat3), 100BASE-T2 (100 Mbit/s over 2 pair Cat3). The section length for a 100BASE-T cable is limit to 100 meters. All are standards under IEEE 802.3. Almost all 100BASE-T installations are 100BASE-TX. Ethernet work over a cable is deemed acceptable for most networks had to be rewiring for 100 megabit speed whether or not there had supposedly been CAT3 or CAT5 cable.

Version of Fast Ethernet over optical fiber is 100BASE-FX, 100BASE-SX, 100BASE-BX and 100BASE-LX10.

The version 100BASE-FX is use 1300nm near-infrared (NIR) light wavelength transmit via to 2 strands of fiber optical, one is for receive (RX) and other is for transmit (TX). The maximum length is 400 Meters for half-duplex connections and 2 KM for full-duplex over multi-mode fiber optical.

The version for 100BASE-SX is use two stands of multi-mode fiber optical to receive and transmit. It is a lower cost alternative, because it use short wavelength. The 100BASE-SX can operate distances is up to 550 Meters. The 100BASE-SX is the shorter wavelength use 850nm and the shorter distance it can support, and use less optical components.

The version for 100 BASE-BX is single strand of fiber optical. the single mode fiber is use along with a special multiplexer splits the signal to transmit and receive wavelengths. The 2 wavelengths is use for transmit and receive is 1310/1550nm. The terminals each side is not equal, as the one transmitting "downstream" use 1550nm wavelength, and other side is "upstream" use the 1310 nm wavelength the distances can be 10 , 20 or 40 KM.

The version for 100BASE-LX10 is two single mode fibers optical. Is also can support 10KM and wavelength is 1310nm. It can describe in IEEE 802.3.

Question 2

C. Differentiate between a hub and a switch. Also use a diagram to show the difference.

Answer 2.C

Label 2.3 is OSI Model for switch and hub

Label 2.3.1 is Micro segmentation (Switch) and no Micro segmentation (Hub)


A hub is rather device in operates on the Physical Layer of the OSI model. Switch is on other hand is more intelligent and it operates on the Data Layer of the OSI model.

A hub will receive the information on one port, information is then broadcast to all ports. This will not be a good plan, if it does not waste bandwidth and causing the collision. Imagine if two computers transmit data at the same time: the impact of information and information package will be destroyed. We will have to relay the data through the process of Ethernet Carrier Sense Multiple Access with Collision Detection, but you can call it a CSMA / CD. In simple terms, it is the protocol we resend the data after a collision occurs.

Collision is clearly a problem with hubs. More importantly the hub is notorious waste the bandwidth. Hubs are operating in half-duplex, the mean is that data can only flow in one direction at a times. As compare with full-duplex where can send and receive the data between two devices at the same times. Since we are operating in half-duplex, the bandwidth will be share between each port on the hubs. For example you have a 20-port hub with 20KB of / s line sharing.  But you can only get 1k-bit / s to each computer on the network.


Switch is operates on the Data Link Layer of the OSI model. This means the switches are quite intelligent than hubs this is because they can route the data in a dynamic level. For example if the information is belonging to computer C, the switch will only send the data to computer C. For example the Label 2.3.1 there are many collision domains for the switch network. If the computer 1 and computer 2 to send data to each other at the same time, for example, you might have a collision. Computer 1 and Computer 3 or 4, however, will not experience the collision process. In the hub network, only have one collision domain. This mean if the first computer wants to transmit a data it can be interrupt by any other computer on the network. Switch can keep track of which computer address are belong to a specific port. If you have information want deliver to computer A, it will only pass thought the computer A port. Micro segmentation allows us to maintain the highest possible bandwidth for each computer. If you have a 20KB of / s line in, each computer can have the full 20KB / s (Note, if two or more computers using a cable at the same time, they must be shared. Nevertheless, it is far better than a hub, it can automatically split the bandwidth of the port, you might not even use!).

Hubs, was chiefly because the price for cheap, easy to install. Fortunately, on the "dumb"devices is that they do not need too much configuration or maintenance.

Switch is more expensive than hubs, but they allow more configuration options. They can be programmed as routers, something hubs can not be completed.

Question 2

Describe frame exchange protocol in IEEE 802.11. Explain how the physical layer specifications for IEEE 802.11 have been issued.

Answer 2.D

The current 802.11 standard defines the "frame" type used in the transmission of data, and the management and control of the wireless link.

A frame is divided into very specific parts and specifications. Each frame has a MAC header, payload and frame check sequence (FCS). Some frames may be not part payload. The first 2 bytes of MAC header is the frame control field, providing detailed information framework. The frame of sub-areas, the area proposed to control the order.

Protocol Version: This is the two representatives in size and protocol version. Current version of the protocol is zero. Other values ​​are reserved for future use.

Type: This is the size of the two bits, to help determine the type of wireless LAN frames. Control, data and management of various frame types are defined in IEEE 802.11.

Sub-Type: This is the 4-bit size. Type and subtype combined to determine the exact frame.

ToDS and FromDS: Each is one bit in size. They indicate whether a data frame is to a distributed system. Control and management frames set these values ​​to zero. All of the data frame will have one bit set. However, in the communication network in the IBSS bit is always set to zero.

More clips: more fragments bit is set to a higher level, the most obvious package is divided, and for all non-final part. Some management framework, and may need to split.

Retry: sometimes need to retransmit the frame, and this has a retry bit is set to a re-send. This helps in the elimination of duplicate frames.

Power Management: Power Management Power Management bits that the sender of the state to complete an exchange. Access points are required to manage the connection will not be set up power-saving bit.

More data: The more data than the buffer frames received for a distributed system. The access point to facilitate is use of the station's power-saving mode. This suggests that at least one station is available and processing of all connections.

WEP: WEP bit after processing a change. This is a framework to switch to one another has been decrypted, or if no will has been one of the encryption settings.

Order: This bit is only set when the "strict orders" transfer method is employed. Frame and the fragments are not always sending the command; it will result in transmission performance.

The next two bytes are reserved in the ID field during this period. This field can take the following three forms: the duration of contention free period (CFP qualified) and the associated ID (aid).

802.11 up to four address fields. Each can carry the MAC address. Address 1 is the receiver, the transmitter address 2, address 3 is the receiver for filtering purposes.

Sequence control field is part of a two-byte message is used to determine the order and the elimination of duplicate frames. The first 4 bits the number of fragments, the last 12 bits are serial number.

An optional two-byte control area is the quality of service added to 802.11.

Frame body field size is variable, from 0 to 2304 bytes, plus any overhead from security package and contains information from the higher level.

Frame Check Sequence (FCS) is the last 4 bytes, in the standard 802.11 frame. Often referred to as cyclic redundancy check (CRC), which allows retrieval framework integrity check? Since the frame will be calculated and sent additional FCS. When a receiver one can calculate the FCS is the framework and compare it to a home. Management framework allows the maintenance of communication. Some of the common subtypes 802.11 include:

Authentication Frame: 802.11 certification began in the WNIC sends an authentication frame of the access point contains its identity. With open system authentication WNIC sends an authentication frame and the only access point responds with an authentication accept or reject the frame itself. With shared key authentication, the initial WNIC sends an authentication request, will receive an authentication framework that contains the text from the access point challenges. The WNIC sends an authentication frame containing challenge text to the encrypted version of the access point. The results of this process determine the WNIC the authentication state.

Association request frame: it is sent from the station to the access point to allocate resources and synchronization. Information about the framework, including support for WNIC data rate and the network SSID station would like to agree. If the request is accepted, the reserved memory of the access point, the establishment of an association ID WNIC.

Association response frame: from one access point to send a station that contains a group to accept or reject the request. If it is an accepted framework will include an associated ID and other information, and supports data transfer rates.

Beacon frames: sent from an access point periodically announce their existence and provide the SSID, and other parameters WNICs range.

Deauthentication frame: terminates the connection from the station hopes to send from another station.

Divorce framework: Send wishes to terminate the connection from the station. This is an elegant way for the access point, to give up the memory allocation and deletion WNIC from the association table.

Probe Request frames: from the bottom station, the requested information to another station.

Probe Response frames: sent from an access point that contains features, supported data rates, receiving Probe Request frames.

Re-association request frame: WNIC sends a request, drop weight combination, from the current range of related and found another access point access point with a stronger signal. The coordinates of the new access points to forward any information that may still be included in the buffer before the access point.

Re-association response frames: sent from one access point to accept or reject the WNIC containing re-association request frame. The framework includes the necessary data is related, such as association ID and supported data rates.

Control framework for the exchange of data frames between stations. Some of the common 802.11 control frame include:

Acknowledge (ACK) frame: the receipt of a data frame, the receiving station will send an ACK frame transmission station, if no errors were found. If the sending station does not receive the ACK frame at a predetermined period of time, the sending station will resend the frame.

Request to Send (RTS) frame: RTS and CTS frame provides an optional collision mitigation plan and the access point hidden station. A station sends a first step in real-time strategy framework for the two shake hands, you need to send data frames.

Clear to Send (CTS) of the framework: a station in response to a RTS frame by frame tour. It provides the required clearance stations send data frames. CTS provides collision control and management, including time value, all other stations are held, and requests the transmission station transmission.

Data frame to carry the package from the website and documents in the body.

In IEEE802 Executive Committee approved two projects to a higher rate of physical layer (PHY) extension to 802.11. The first extension, IEEE 802.11a, and defines the requirements for a physical layer operating in the 5.0 GHz U - NII frequency and data rates from 54 Mbps to 6 Mbps of. In the second extension, IEEE 802.11b, and defines the physical layer specifications set work in the 2.4 GHz ISM band up to 11 Mbps. Both physical layers are definitions of operation and the existing MAC. The physical layer of the 802.11 consistent three wireless data exchange method, they if infrared (IR), frequency hopping spread spectrum (FHSS) and direct sequence spread spectrum (DSSS). 802.11 Wireless LAN radio operators in the 2.4GHz (2.4 è‡³ 2.483 GHz) unlicensed radio frequency (RF) band. Isotropic maximum transmit power allowed in this band in the U.S. Federal Communications Commission 1WT type, but the equipment is usually limited to 100mWt 802.11 values.

In the 802.11 physical layers is divided into the physical layer convergence protocol (PLCP report) and physical medium dependent (PMD) sub layer. Be reported in the PLCP / parse the data unit to send / receive using a variety of 802.11 media access technology. PMD of the data transmission / reception and modulation / demodulation of direct access, under the guidance of air PLCP reported. 802.11 MAC layer is subject to the great extension of the nature of the media. For example, it implements a second layer of relatively complex broken PDU.

For the infrared (IR) for wireless are not been acceptable by public, this is because there are no successful commercial implementations on 802.11 IR technology.

For the FHSS is use Radio frequency for transmission to the rapid change in the transmission process. The first short rupture is transmitted at one frequency, the second on another frequency, and so on. The amount of time spent at a particular frequency known as the residence time, change the frequency hopping sequence is known. According to FCC regulation, all in the 900 MHz band FHSS systems hop must be 50 channels, can not spend more than four into a second frequency in a transmission time every 20 seconds. FHSS system in the 2.4 GHz band to jump between the 15 channels, the maximum output power can not be more than 124 mW. FHSS is not widely used in wireless LAN system, but the Bluetooth does not use it in the 2.4 GHz frequency. Bluetooth change frequency 1600times.

For the Direct Sequence Spread Spectrum (DSSS) is a inflection technique. Like all the other spread-spectrum technology, the transmitted signal takes up more bandwidth than the information signal is modulated. This technique makes use of pseudo-random string of consecutive PN code symbols called "chips." These chips have a shorter duration than one bit of information, because this method of information transmission regulatory sequences of the chip. Also uses a spread spectrum signal sequence structure, in which the chip is generated by a transmitter called the a priori receiver. The receiver can use the same PN sequence offset the impact of the PN sequence in the received signal to reconstruct the information signal.

Makes use of pseudo-random sequence spread spectrum transmission 1 and -1 values ​​and the data being transmitted multiplied by a "noise" signal. This will create a transmission frequency is much higher than the original signal, which spread the energy of the original signal into a wider band. This will create a similar "white noise" or static, the only real difference is that with the actual static, the receiver can extract meaningful data is multiplied by the same pseudo-random sequence. This process is called "to spread."

For the cancellation spread to work, send and receive sequences must be synchronized. This requires the serial receiver and transmitter synchronization sequence through some form of search time. However, this obvious shortcoming, it is obvious benefits: if the sequence number of transmitters synchronized relative to each other so that they must be synchronized between the receiver can be used to determine the relative time, which in turn can be used to calculate the receiverposition, if the transmitter's position is well known. This is the basis for many of the satellite navigation system.

The resulting effect of channel noise ratio is called processing gain. This influence can be through the use of larger and longer PN sequence chips per bit more, but the actual physical device used to generate the PN sequence impose limitations to the processing gain.

If do not want to send the transmitter, but in the same channel using a different PN sequence (or no order) in the communication process to achieve the results did not get a signal. This effect is based on Code Division Multiple Access (CDMA) spread-spectrum property, which allows multiple transmitters to share the same channel within the limits of cross-correlation properties of PN sequences.

As this description suggests that the emission waveform of a conspiracy to have a general bell-shaped envelope concentrated in the carrier frequency, just like a normal morning transmission, but the increase in noise to a wider distribution than a morning transmission.

In contrast, pseudo-random frequency hopping spread spectrum re-tune the carrier, rather than pseudo-random noise added data, results in a uniform frequency distribution, the width of the output range from pseudo-random number generator.

The advantages of spread-spectrum resistance to intentional or unintentional interference, sharing a single channel for multiple users, and reduce the spread of opportunities, one will be blocked.

Question 3

a. Explain circuit switching and the phases used in it with the help of a diagram.

Answer 3.A

Label 3.1 Circuit Switching

Circuit-switched (Circuit Switching) is a relative concept of packet switching. Circuit switching requirements must first establish a connection between the two parties in the communication channel. After the connection is established, the two sides of the communication activities can begin. Both sides need to communicate the message through a good connection has been established for delivery, but the connection has-been maintained until the end of the communication of both sides. Communication activities at a time throughout the process, the connection will always occupy the beginning of the connection is established, communication system resources allocated to it (channel, bandwidth, time slot, code, etc.), which reflects the difference between circuit switching on the essential characteristics of packet switching. In this networking, the connection between two devices is called the circuit, which issued throughout the communication. Nature of the information, the circuit is the maintained by the network. The circuit may also be able to be fix one that is always present, or it may be a circuit that is created on an as-needed basis. Even though many potential paths through the intermediate device may have communication between two devices, only one will be used for any particular dialogue. In the circuit-switched network, communications can occur between two devices in front of the circuits. This is shown as a thick blue line of the pipeline data from the device device A to Device B and matching purple line from device B back to device A. Once established, all communication between these devices in this circuit, even if there are other possible ways can be envisage through the data between devices on the network. The most typical example of the circuit-switched is telephone network system. When you call, they answer, you establish a circuit connection between the data you can, if needs a steady flow. This circuit functions the same way, no matter how many intermediate devices are used to your voice. You can use it, as long as you need it, and then terminate the circuit. The next time you call, you will get a new circuit, which may be use a different hardware than the first circuit, depending on what can be time in the network.

Question 3

b. What is packet switching? Describe Datagram and Virtual circuits in packet switching with the diagrams

Answer 3.B

Label 3.2 Packet Switching

In computer networks and communications, packet switching is a communications paradigm, group (message or message fragments) in a single route between nodes, no previously established communication path. Packet switching data communications is a new and important concept, is now the world's data and voice communications in the most important foundation. Previously, data communication is based on the idea of â€‹â€‹circuit switching, as in traditional telephone circuits, in a call needs to occupy a proprietary circuit, communication in both ends of the circuit. In this network type, there is no specific path is used for data transmission. Instead, the data is chopped into small pieces called packets and sent over the network. Packets can be routed, combined or distributed as required to their final destination. At the receiving end, this process is the opposite, read data packets and re-assembled into the form of raw data. Packet switching network similar to the postal system more than its telephone system (albeit not perfect.) An example is shown in Label 3.2. In the packet-switching network, no circuit is establishment prior of transmission the data between devices. Blocks of data, even from the same document or communication, can take any number of paths, because it's journey from one device to another. Compare this with Label 3.1. In packet switching, a system can be assembled into packets of data using a communication link with more than one machine communication. Not only the link is shared, and each packet independent of other packets can be routed. This is the main advantage of packet switching.