Comp Net
146 flashcards · Other · by Minh Đức
what is the general purpose of Computer Network ?
enable communication among the participants
for setting up and running a computer network, which components are required
>= 2 computers with network services running transmission medium network protocol
how computer networks can be distinguished by their dimension?
Personal Area Network or Body Area Network Local Area Network Metropolitan Area Network Wide Area Network
Describe PAN / BAN property range technologies
network of small devices range: a few meters technologies: bluetooth, WLAN, USB
Describe LAN property range cover technologies
local network range:(500-1000m) apartment, building, university campus technologies: ethernet, WLAN
Describe MAN property range cover technologies
connect LANS range: 100km, cover a city fiber-optic cable
Describe WAN property range cover technologies
connect several networks 1000km, large geographic area inside a country or continent Asynchronous transfer mode, ethernet
what is unicast?
one to one communication one host sends information to 1 other host
what is broadcast
one to all one host sends information to all the hosts in the network
what is multicast?
one host sends information to all hosts in a given group
what is anycast?
one host sends information to one host in a given group
what does connection-orientation means?
network services may operate connection-oriented or connectionless
3 types directional dependence of data transmission?
simplex: only one side of the channel can send data full-duplex: both side of the channel are able to send half-duplex: both side can send but not simultaneously
what is bandwidth (data rate)
how many bits can be transmitted within a period via the network
what is throughput
actual achieved data rate
what is goodput
actual data rate that user benefits from
what is latency
time that a message needs to travel from one end to the most distant end of the network
formular for latency
latency = propagation time + transmission time + waiting time
formular for propagation time
t = distance / c * velocity factor
formular for transmission time
t = message size / bandwidth
formular for volume of network
V = propagation time * bandwith
what are the layers of TCP/IP model and the protocol of each layer ?
application layer (HTTP, SMTP, POP3,...) transport layer(TCP,...) internet layer (IPv4, IPv6,...) link layer (Ethernet, WLAN)
what are the layers of hybrid model
application transport network data link physical link
each layer will add additional information as header to the message
which layer will also add trailer at the end of the message
link layer
what does the physical layer do
transmit the 1s and 0s, convert data into signals
in physical data, at the sender site: signals are modulated onto the medium At receiver site: Signals are demodulated from the medium
what does the data link layer do
Ensures error-free data exchange of frames between devices in physical networks
what happen at the sender site of the data link layer
pack the network layer packets into frames and send them via physical network
what happen at the receiver site of the data link layer
Identifies frames in the bit stream from the Physical Layer
what does network layer do
Forwards packets between logical networks (over physical networks)
what happen at the sender site of the network layer
Packs the segments of the Transport Layer in packet
what happen at the receiver site of the network layer
Unpacks the packets in the frames from the Data Link Layer
what does the transport layer do?
Transports segments between processes on different devices via so-called end-to-end protocols
what happen at the sender site of the transport layer
Packs the data of the Application Layer into segments
what happen at the receiver site of the transport layer
Unpacks the segments inside the packets from the network layer
what does the application layer do ?
Contains all protocols, that interact with the application programs (e.g., browser or email program)
in which layer defines the IP address
network layer
Addresses processes with port numbers in transport layer
which layer ?
Transport protocols implement different forms of communication in transport layer Connectionless communication, typically UDP (User Datagram Protocol) in TCP/IP networks Connection-oriented communication, typically TCP (Transport Control Protocol) in TCP/IP networks
which protocol
Today, Ethernet (1-10 Gbit/s) with Switches (star topology) is the standard for wired LAN Connecting Hubs and Switches implements a tree topology, if there are no loops in the cabling Cell topology is the standard for wireless networks Mesh topology is one possible use case of wireless networks and it is the logical topology between routers
Why are the Presentation Layer and the Session Layer not intensively used?
The functionalities, which are intended for the Session Layer and Presentation Layer, are now part of protocols and services in the Application Layer.
Why is the hybrid reference model closer to reality, compared with the TCP/IP reference model?
distinguishes between the Physical Layer and Data Link Layer and it does not subdivide the Application Layer. take the advantages of TCP/IP and OSI models
relation between bit rate and symbol rate
The bit rate depends on the bandwidth of the communication channel and the number of bits per symbol
Under which circumstances are symbol rate and bit rate equal?
If the carrier signal has only two states, then only one bit of data (i.e., a 0 or 1) can be transmitted in each symbol.
Explain why computer networks require line codes.
Computers are digital machines. Transmission mediums work analogously. The line codes specify the conversion of binary data into signals (encoding).
Many different line codes exist. Explain why it is impossible to use one single line code for every network technology.
Different transmission mediums are used for computer networks. Different numbers of signal levels are used.
Explain the way Non-Return-To-Zero (NRZ) works
It represents logical 0 and 1 is by using different voltage levels.
Name the two problems that can occur when NRZ is used to encode data.
Baseline Wander and Clock Recovery
what is baseline wander problem
When transmitting a long series of logical 0 bits or 1 bits, the average can shift so much, making it difficult to detect a significant change in the physical signal
what is clock recovery problem
the processes for encoding and decoding run on different computers, they need to be controlled by the same clock. In each clock cycle, the sender transmits a bit and the receiver receives a bit. If the clocks of sender and receiver drift apart, the receiver may lose count during a long sequence of logic 0 bits or 1 bits.
how to prevent baseline wander
when using a line code with 2 physical signal levels, the usage of both signal levels must be equally distributed.
how to prevent clock recovery problem
increase the number of guaranteed signal level changes
Name at least 5 line codes that use 2 signals levels
NRZ, NRZI, Unipolar RZ, Manchester, Manchester II, Differential Manchester.
Name at least 3 line codes that use 3 signal levels.
MLT-3, RZ, AMI,
Which line codes ensure a signal level change for each logical 1 bit?
NRZI, MLT-3, Unipolar RZ, AMI
Which line codes ensure a signal level change for each transmitted bit
RZ, Manchester, Manchester II, Differential Manchester
Why do not all line codes ensure a signal level change for each transmitted bit?
Lack of efficiency
Which line codes ensure that the signal levels are equally distributed?
AMI, Manchester, Manchester II.
Why is it important for the receiver of signals, which are encoded according to the Differential Manchester Encoding, to know the initial signal level?
Depending on the initial signal level, two signal sequences, inverse to each other, are possible.
What is a scrambler?
A scrambler is a device, which modifies a data stream according to a simple algorithm in a way that it is easy to reverse.
Why are scramblers used?
When the AMI line code is used, clock recovery is impossible for the receiver, when series of logical 0 bits are transmitted. In AMI case, scramblers are used, to interrupt long series of logic 0 bits. This makes the clock recovery for the receiver possible.
All line codes have drawbacks. What can be done to avoid the problems, that can result from these drawbacks?
Modern network technologies encode the bit stream first with a line code that works efficient on the one hand, but also ensures clock recovery and avoids baseline wander. These encodings improve the bit stream in a way, that a further encoding with the line codes NRZ, NRZI and MLT-3 does not result in any problems. Examples of line codes, which improve the bit stream first, are 4B5B, 5B6B and 8B10B. These line codes encode fixed-size input blocks into fixed-size output blocks.
How is the efficiency of a line code calculated?
Efficiency = ratio of bit rate (payload in bits per time) and baud rate (signal changes per second).
Taking the classical telephony example: How often should the system sample the signal?
The bandwidth of the POTS is given by the voice spectrum → 3100 Hz. Hence, the required sampling frequency is 6200 Hz = 6.2 kHz
formular for sampling frequency
fs = 2*f
formular for maximum data rate
2 ∗ H ∗ log2(V ) V : number of different symbol values H: the channel bandwidth in Hertz (Hz)
in which case, the bit rate < symbol rate
If a line encoding with less than 100% efficiency, e.g., Manchester Code.
Why can a symbol not carry an arbitrary amount of bits?
Because of the noise → upper bound given by the Shannon-Hartley theorem.
formular for SNR
maximum data rate = H ∗ log2(1 + S/N)
which cable types can be used for Ethernet?
coxial, twisted pair and fibre optic
what is the main difference between twisted pair and coaxial?
coaxial consists only one wire => half duplex twisted pair => full duplex
which type of twisted pair cable should be used to connect 2 computers directly
traditionally, crossovers cables but nowadays modern computers can be connected via patch cables
Name the advantage of the fiber optic cables over copper wires
higher transmission rate less noise
what is transmission range
all devices in transmission range are typically able to correctly decode a received signal
what is interference range
devices in interference range cannot distinguish the sent signal from the background noise
challenges for wireless network that do not occur in wired network
reflection/multipath fading and hidden terminal problem
Directed wireless transmission eliminates some of these issues and can achieve higher transmission ranges. Explain why many wireless networks still work with undirected (omnidirectional) transmission systems
WLAN access point provides network access for multiple user undirected transmission allow users to move freely
electromagnetic wave is affected by attenuation over the distance, why is this more prominent in wireless network
the energy of radio waves is absorbed by different materials => signal weakening is less predictable
Metal shielding protects against external interference. Hence, would it be a good idea to use metal tube for a directed radio transmission? Why or why not?
because of the reflection -> multipath fading -> faraday's cage
name 2 popular technologies in school, company,....
ethernet and WLAN
which communication model is used for the school WLAN network?
infrastructure model
in which frequency spectrum do WLAN networks operate
2.4GHz, 5GHz, 6GHz, 60GHz
Network 5 and Network 6 can be identified as IEEE 802.11b networks. Which data rate can be expected?
50% of 11Mbits/s => 5.5 Mbits/s
what is the purpose of bridges in computer network
for connecting different physical networks
What is the advantage of learning bridges in contrast to "dumb“ bridges?
Learning Bridges learn which network devices are accessible via which port and therefore decrease unnecessary traffic.
What information is stored in the forwarding tables of bridges?
which network devices are accessible via which port
Why do bridges try to avoid loops?
Loops can cause malfunctions and reduce the performance of the network or even lead to a network failure
What is the selection criteria for determining, whether a bridge becomes the root bridge?
root ID and MAC address
The format of what addresses is defined by Data Link Layer protocols?
physical network address (MAC address)
Who receives a frame with the destination address FF-FF-FF-FF-FF-FF?
This address is the MAC broadcast address. Every participant in the local/physical network receives this frame.
What is MAC spoofing?
change MAC address via software to fake its identity
Most data link layer protocols put the CRC in the end of a frame (trailer) rather than in the beginning (header). Why?
The CRC is computed during transmission and appended to the output stream as soon as the last bit goes out onto the wire. If the CRC were in the header, it would be necessary to make a pass over the frame to compute the CRC before transmitting. This would require each byte to be handled twice—once for checksum and once for transmitting.
Why do computer networks use protocols for media access control?
When a shared transmission medium is used it coordinates the access to this medium and shall avoid collisions
Describe the differences between deterministic and non-deterministic media access control
In a deterministic MAC the data rate and latency for each station can be predicted because the resources are pre-allocated, a non-deterministic MAC follows a best-effort principle.
Which media access control method is implemented by Ethernet?
Non-deterministic media access control
Which media access control method is implemented by Token Ring?
Deterministic media access control
Which media access control method is implemented by WLAN?
Non-deterministic media access control
What is the advantage of the media access control method of Token Ring in contrast to the media access control method of Ethernet?
In Token Ring, we can predict the waiting time and amount of data that can be transmitted. For Ethernet, all participants access in direct competition.
Why do Ethernet and WLAN use different media access control methods?
With wireless networks, it is not guaranteed that all stations can detect all collisions. In wired networks with a shared transmission medium, each participant receives the transmissions of all other participants.