Data Link Layer
amplifier
A device used to boost the strength of a signal. Amplifiers are spaced at intervals throughout the length of a communication circuit to increase the distance a signal can travel. See also repeater.
repeater
A device used to boost the strength of a signal. Repeaters are spaced at intervals throughout the length of a communication circuit.
hub polling
A type of sequential polling in which the polling device contacts a terminal, that terminal contacts the next terminal, and so on, until all the terminals have been contacted.
Hexadecimal
A way to represent #'s using 16 digits
What is a frequent cause for cross-talk and intermodulation noise?
Improper multiplexing
The physical layer provides data transfer of bits across the link t/f
True
The source usually gets the destination IP address from a Domain Name Server (DNS) (t/f)
True
What are parity codes?
Simplest error detecting code. How many 1 bits are in the data? - an odd number = parity 1 - an even number = parity 0 Transmit the parity with the data, check at the receiver. This detects all single bit errors.
Unicast
Single host address
Which of the following is/are identified by ARP? Congestion in network Errors in packet Size of each address Hardware Type Protocol Type
Size of each address Hardware Type Protocol Type
Type
Smaller than 0x05EE
Switches
Star topology, default is Ethernet, only forward frames to destination hosts
Types of Ethernet MAC address
Unicast, broadcast, multicast
Switching methods
Store-and-forward, cut/through, fragment-free
What is link local?
The address is unique amongst hosts connected to a link, this is often sufficient.
sliding window
a 'window' works to cover the message to indicate that is waiting, and slides when ready
Frame Chek Sequence
a 4 byte (32bit) number that represents a checksum value for the entire frame.
frame
a block of data in which all the letters in one group of data are being transmitted at one time in a synchronous transmission
Media Access Control
controls which device transmits and when - important on: multipoint (shared) circuits half-duplex point-to-point circuits
-inexpensive -easy to install -has low resistence to electrical currents
copper media is....
The frame payload is
data
The tailer contains
error detection and frame stop
burst error
errors appear in more than 1 data bit that is changed by the error causing the condition where errors are not uniformly distributed in time
Hubs
extend the distance that ethernet cabling can reach by regenerating the frame and retransmitting it (like a redstone repeater in minecraft)
Layer 2 addresses are used for
only local media
Controlled access
only one station transmits at a time, devices wanting to transmit must wait their turn
Ethernet frame fields
preamble and start frame delimiter, destination address, source address, type, data, FCS
ARQ:
process of requesting that data transmission be resent
Error Correction
should know number of bits corrupted and location in the message: "Not Simple"
SFD Start Frame Delimiter
signals to the receiving device that the preamble is over and that the actual frame contents will now follow.
Layer 1 elements
signals, bit streams that travel on the media, physical components that put out signals, and various topologies
elevators transformers fluorescent lighting lightning storms
sources of EMI
microwave ovens portable phones
sources of RFI
transmission efficiency
total number of information bits divided by the total number of bits (info + overheard) in transmission
throughput
total number of information bits received per second after taking into account overhead bits and need to retransmit frames containing errors
Throughput
total number of information bits received per second, after taking into account the overhead bits and the need to retransmit frames containing errors
1. Ethernet 2. 802.11 wireless 3. point-to-point protocol (PPP) 4. HDLC 5. Frame Relay
what are 5 common data link layer protocols?
1. frame start 2. addressing 3. type 4. control
what are the four parts of a header?
1. Header 2. Data 3. Trailer
what are the three basic parts of a data link layer frame?
1. error detection 2. frame stop
what are the two parts of a trailer?
Frame check sequence (FCS) Cyclic redundancy check (CRC)
what are the two types of error checking?
1. Technologies for signaling and detecting network signals 2. Problems with physical media
what factors determine the practical bandwidth of a network? (hint 2)
1. latency 2. amount of traffic 3. type of traffic
what factors influence throughput? (hint 3)
used to verify the integrity of a frame
what is the purpose of error checking?
odd parity
when there are 2 1's in in a 8-bit ASCII, the parity bit is set to 1 to make it an odd number of 1s
even parity
when there are 3 1's in in a 8-bit ASCII, the parity bit is set to 1 to make it an even number of 1s
because one is sending and the other one is recieveing
why are two strands necessary for full duplex?
What are the characteristics of global scope addresses?
✔ Simpler to implement if devices can move, since don't need to change address when connected to a different link. ✘ Exists some privacy concerns.
Electromagnetic interface (EMI) of Radio frequency interface (RFI)
Signals can distort and corrupt the data signals being carried by copper media
Truck example:
The data link layer gets data, puts it in an envelope, and tells where it should be sent. At home, if somebody tells you to send a letter and they give you the letter, then you put it in an envelope, put the address on it, and send it. After that, the truck picks it up and delivers it (the physical layer).
misalignment
The fiber- optic media are not precisely aligned to one another when joined
Attenuation
The loss of power a signal suffers as it travels from the transmitting computer to the receiving computer. Increases as frequency increases or as the diameter of the wire decreases.
End gap
The media does not completely touch at the splice or connection
Negative acknowledgement (NAK)
The message contained an error
What is the tradeoff in adding error correcting code?
The more complex the error correction, the greater the ability to correct multi-bit errors, but the more data we have to add.
Network topologies
The way nodes are connected to other nodes
Explain how the link utilisation in sliding window protocols is linked to the window size w and the bandwidth-delay product BD. Assume the bandwidth delay product is expressed in frames per second. (6 marks) 2015
The window width w should be 2BD + 1, that is, the number of frames that can be sent over the channel during a round trip to the receiver plus one (the first acknowledgment will reach A after 2BD + 1 frames are sent by A). The relationship with the link utilisation η is n =< w/ 2BD + 1
How many bytes are Ethernet MAC addresses?
6
How many bits is an Ethernet address?
6 bytes - 48 bits (globally unique addresses).
Ethernet host MAC address
6 bytes, 48 bits, OUI and Vendor
Preamble
7 bytes of 10101010 followed by 1 byte of 10101011, alerts other modes frame is about to be sent, provides clock synchronization
Preamble
8 bytes (64 bits) long and can itself be split into 2 sections
What is encoding in NRZ? How does it work?
A common "language" that is used to send information from the sender (encoder) to the receiver (decoder). Low level for 0. High level for 1.
media access control
A data link layer protocol that defines how packets are transmitted on a local area network. See also CSMA/CD, token bus, and token ring.
Full mesh
Each node has a point to point connection to every other node
What do multi-access links require?
Host addresses, to identify senders and receivers.
T/F: multipoint configurationns share the same circuit
True
Receiving side?
- looks for errors, does flow control - extracts data, passes to network layer
What do the data link protocols differ by?
- message delineation - frame length - frame field structure
Throughput
- more complicated than simple protocol efficiency because it depends on the retransmission rate, transmission rate and delay - TRIB
What is async Efficiency?
= 8/(8+3) = 0.73 = 73%
0+0
=0
Random Access Protocol
- no prior coordination or channel assignment - when node has packet to send - specifies how to detect collision and how to recover from them
Taking Turns
- nodes take turns - nodes that need to send more, can take longer turns
1+1
=0
0+1
=1
1+0
=1
What are the sublayers of the Data Link Layer?
Logical Link Control sublayer (LLC) Media Access Control sublayer (MAC)
What is Manchester encoding?
A 0 is expressed by a low to high transition. A 1 is expressed by a high to low transition.
acknowledgment (ACK)
A character indicating a positive acknowledgment that a message has been received correctly.
Rollover
A cisco proprietary cable, it connects to console port on networking equipment
Stop and Wait (half duplex)
(1) Sender sends a message and waits for ACK and then sends the next message (2) Reciever gets the message and sends an ACK or negative ACK when its ready to recieve more packets
Multiple access links
- Point to point -- Dial up - Broadcast -- Shared RF; Satellite, wifi. -- Shared wire
Error Detection - Parity Check
- 1-bit check value - Based on the number of 1's in the message -- Even parity: number of 1's remains even -- Odd parity: number of 1's remains odd - Simple, but only detects 50% of errors
Error Detection - Checksum
- 1-byte (typically) check value - Checksum algorithms vary in the creation of check values - Detects 95% of errors
Switches
- Are nodes - Transparent - Hosts communicating are not aware of a switches presence - Operates in the link layer, used to connect two or more network segments and regulate traffic. - Self learning - hosts have dedicated connection to switches - Full duplex - Link layer (ethernet) protocols used on each link but has no collision -- each link has its own collision domain - Switch forwarding table -- attr(Mac address, interface line, time to live) - records sender location pari in table
Data Link Protocol Classifications
- Asynchronous transmission - Synchronous transmission
What is framing?
- At transmitter, link layer encapsulate datagram into frame, adding header (ex. address) trailer
How does the data link correct these problems?
- Breaks raw bit stream into frames. - Transmit and repair individual frames. - Limit scope of any transmission errors.
What are the MAC protocols categories?
- Channel Partitioning - Random Access - Taking turns
Even Parity
- Check that the number of 1's in the data are even, if not the parity bit becomes a 1. -> transmit data - if the receiver detects that the number of bits, including the parity, are odd then there is an error.
Odd Parity
- Check that the number of 1's in the data are odd, if not the parity bit becomes a 1. -> transmit data - if the receiver detects that the number of bits, including the parity, are even then there is an error.
Multiple access protocol
- Co-ordination must use same channel as communication. - Distributed
CDMA
- Code division multiple access - assigns a different code to each device - with proper code design, nodes can send simultaneously, yet the receivers can correctly decode the information -early used in military systems, now used in civilian applications
Channel Partitioning
- Divide channel into smaller pieces - assign a piece to each node ex. time slots, frequency
Sending NIC Adapters
- Encapsulate data in a frame - Adds headers -- Error checking bits -- rdt -- Flow control
Frame
- Encapsulates a datagram - All data from lower level protocols -- Including lower level headers -- And its own header --- protocol identifier --- tag control information -- And its own trailer ---error detection bits, CRC, parity, etc. - Uses MAC addressing
FDMA
- Frequency division multiple access - the broadcast channel is divided into different frequencies and each device is assigned to a different frequency
IP
- Hierarchical assigned by isp's local router and global isps - Routing calculations are based on destination address IPv4 - 32 bit - used in network layer linking IPv6 - 128bit - used in network layer linking
Nodes
- Hosts and routers - Exist physically in space - All nodes contain an NIC -- Except Switches
Data Link Protocols - Ethernet
- IEEE 802.3 standard and Ethernet II - Most widely used LAN protocol - Uses contention media access control Ethernet II Frame: Preamble & Delimiter (8 bytes) + Destination Address (8 bytes) + Source Address (8 bytes) + Type (2 bytes) + Data (46 - 1500 bytes) + CRC (4 bytes) | Interframe Gap ( 12 bytes) | = 38 Total Bytes of Overhead
MPLS vs IP
- IP route calculation based only on destination address - MPLS based on source and destination address -- Also has dedicated backups
Link Layer Services
- Identifies existence of a connection - Verifies if a connection is free to access - single bit error detection -- Single bit parity - Single bit error correction -- 2D parity - Multi bit error correction -- CRC - Flow control -- Allows everyone on the network to have a chance to transmit - Adjacent node traffic control - Duplex - Half-duplex - Full-duplex
Error Correction - Retransmission - Stop-and-Wait ARQ
- If no ACK or NAK, sender retransmits frame after "timeout" - If no ACK, sender resends data, receiver sends ACK and deletes duplicate frame
Receiving NIC Adapters
- Looks for errors - Extracts datagram and passes to higher layer protocols
What is addressing?
- MAC address usually locally on a link
Sources of Network Errors: Line noise and distortion
- Major reason for errors and caused by several sources - More likely on electrical media - Undesirable electrical Signal - Degrades performance of a circuit - Manifestation -- Extra bits -- "Flipped" bits -- Missing bits
Random Access
- No pre-allocation - allows collisions, then recovers from collisions ex. channel not divided
Token passing
- Nodes with a token are the only nodes in the network that are allowed to use the link - token has a ttl - when a node has finished transmitting it passes the token to another node on the network - Token passing has token overhead -- Nodes will wait forever to transmit if a node fails to relinquish control of a token
Error Correction - Error control in practice
- On wired connections, errors are quite rare - Most data ink layer software today does not correct errors only detect them and discard frames with errors - error correction must then be done at a higher layer (Transport)
Error Correction
- Once detected, errors must be corrected - Error Correction Techniques: Retransmission (or backward error correction), Forward error correction
ODMA
- Orthogonal frequency division multiple access - assign a subset of orthogonal subcarriers to each user
Taking turns
- Polling - Token passing
Transmission Efficiency
- Protocol design contributes to network efficiency = Total number of information bits (i.e., bits in the message sent by the user)/ total bits in transmission (i.e., information bits plus overhead bits)
2D parity
- Put all data segments into a matrix. - Horizontal parity check - Vertical parity check - if an error is detected then parity will be off in two dimensions and single bit error can be corrected
MAP Protocols
- Random access - Taking turns - channel partitioning
ALOHA
- Random access protocol - When information is received from the higher protocol layers transmit it immediately - Huge chance of collision - maximum successful usage of the line is 18%
Slotted Aloha
- Random access protocol - all frames in traffic through the node are the same size - Allocated time frame for nodes to begin sending - Nodes are synchronized - All nodes aware of collisions occurring - best case usage of line 37% - when nodes' transmissions collide then nodes send jamming signal and cancel transmission, they wait until next sequence and then have a probability to send again.
Error Detection
- Receivers need to know when the data transmitted is not correct - Add "check value" (error detection value) to message -- [Message + Check Value] - Check value produced by mathematical formula - Both sender and receiver calculate check value - Sender tests whether the check values match
Forward error correction
- Receiving device can correct messages without retransmission
Retransmission ( or backward error correction)
- Retransmission is simple and most common - Automatic Repeat reQuest (ARQ) -- Stop-and-Wait ARQ -- Continuous ARQ - This can also provide flow control by limiting the number of messages sent
What does a frame consist of?
- Start code: used for synchronisation and timing recovery. - Header: contains host addresses and control info. - Data: the network layer protocol data. - Error detection.
Data Link Protocols- Synchronous Data Link Control (SDLC)
- Synchronous bit-oriented protocol developed by IBM - uses bit stuffing (zero insertion) to overcome transparency problem Format: Flag (8 bits) + Address (8 bits) + Control (8 bits) + Data (variable) + CRC-32 (32 bits) + Flag (8 bits)
What is the most common end-user implementation of the data link layer?
- The ethernet - most corporate desktops
TDMA
- Time Division Multiple Access - Each station gets fixed length time slot ex. (length = maximum packet transmission time) in each round
What is the function of the Data Link layer?
- Transfer data between individual links that makeup the end-to-end transmission path - Detect and possibly correct errors that may occur in Physical Layer
Contention Access
- Transmits whenever circuit is available with no centralized control - common in Ethernet LANs - devices must be "polite" and follow these steps; 1. "Listen" for traffic 2. If another device is transmitting, wait to transmit 3. Otherwise, transmit (and keep listening) 4. If another device begins to transmit, stop and wait
Encapsulation
- Transportation and application layers are implemented at end devices (end to end)
Random access protocols
- When a node has a packet to send, attempt to transmit the node at band rate. - No co-ordination among nodes - Nodes don't attempt to avoid collisions but try to recover from them -- collision detection -- collision correction
What is the most common wireless link layer
- Wifi
There is reliable delivery between _______
- adjacent nodes
Data Link Protocols - Point-to-Point Protocol ( PPP)
- common WAN protocol Format: Flag (1 byte) + Address (1 byte) + Control (1 byte) + Protocol (2 bytes) + Data (variable) + CRC (2 or 4 bytes) + Flag (1 byte)
Controlled Access
- common in wireless LANs - Access Request -- each device must get "permission" to transmit, similar to raising a hand - polling --roll-call polling ----central device (controller) determine which devices can transmit ----Each client is checked periodically t see if it needs to transmit -- Hub Polling (token passing) ---- One device begins the poll and then passes it to another device until it reaches them all
What is Medium Access Control,( if shared medium)?
- control who uses channel
Synchronous Transmission
- data sent in a large block called a frame - includes addressing information - includes synchronization characters to let the receiver know when data transmission begins - Example protocols: SDLC, HDLC, Ethernet, PPP
Error Correction - Hamming Code
- each data bit figures into three EVEN parity bit calculations - If any one bit (parity or data) changes --> change in data bit can be detected and corrected - only works for one bit errors
Random Access: Slotted ALOHA
- fix the starting/ending times to specific values - say t, t+F, t+2F and so on -this reduces the number of collisions - node has frame to send, waits till beginning of next slot and transmits the frame - If collision occurs no acknowledgement is received
Hardware implements both of what functions?
- hardware implements both the link layer and physical layer
What is one similarity between TDMA and FDMA?
- in both the channel bandwidth is divided evenly among devices
Error Correction - Forward Error Correction
- includes a certain level of redundancy in transmitted data so that receiving device can correct errors - does not require transmission - used only when retransmission is impossible, very costs or time consuming (e.g. satellite connections)
How can we increase transmission efficiency?
- increase data bits - decrease overhead bits
Two or more simultaneous transmissions by nodes can result in what?
- interference - Collision if node receives two or more signals at same time - All frames involved in collisions are lost
Asynchronous serial transmission (async)
- old protocol (e.g. used in teletype) - Transmits one character at a time - Delineation indicated by start and stop bits Format: Startbit (1) + Message (7 or 8 bits) + Parity bit (1) + Stopbit (1)
(Reliable delivery between adjacent nodes) What is flow control?
- pacing between adjacent sending and receiving nodes
What are the two different types of links?
- point to point - broadcast
(Reliable delivery between adjacent nodes) What is error correction?
- receiver asks for retransmission if error occurs
(Reliable delivery between adjacent nodes) What is error detection?
- receiver detects presence of errors
Multiple Access Protocols (Medium Access Control)
- regulate the transmission of devices into a single shared broadcast link
Error Correction - Retransmission - Stop-and-Wait ARQ
- sender does not wait for ACKs before sending more data - "Sliding window" is number of frames allowed to be unacknowledged by receiver - Agreed upon by sender and receiver
What is a broadcast link?
- single broadcast channel connecting multiple senders and receivers - ethernet, wifi
What is a point-to-point link?
- single sender at point of the link and single receiver on the other end
Where are data link functions implemented?
- they are implemented in the hardware adaptor aka network interfacing card -ethernet card
Random Access: ALOHA
- transmit a link level frame whenever you want - wait for an acknowledgement (ACK) - If there is a collision, wait till a random amount of time and retransmit
Error Detection - cyclic redundancy check (CRC)
- treats message as a single binary number - divides by a preset number - uses remainder as a check value - Preset number is chosen so that remainder is the correct number of bits - Modes: -- CRC - 16 (~99.998% error detection rate) -- CRC - 32 (>99.99999% error detection rate)
When are MAC protocols needed?
- when using a single shared broadcast link
Examples of Network Error Example:
-Bit value change during transmission -Controlled by network hardware and software
What are the network error categories ?
-Corrupt -Lost Data
Polling
-Master node controls slave nodes transmissions -Slave nodes have no processing and just follow instructions -Single point of failure for the connection if master node drops - sub par latency - Violates decentralized node transmission ideal.
Two types of Polling
-Roll Call Polling -Hub Polling (AKA Token Passing)
What are the main AQR Protocols?
-Stop and Wait (half duplex) - Continuous (full duplex)
What are major Controlled Access Methods?
-access request -polling
synchronous transmission
-all the letters of data in one group of data are transmitted at one time as a frame -computer will save all of the keystrokes typed by the user and transmit then only when the user presses a 'transmit' key -start & end key of the entire frame must me marked rather than for eahc letter
asynchronous transmission
-often referred to as start-and-stop transmission -transmitting computer con transmit a character whenever it is convenient and the receiving will accept the same way -typically used on point-to-point full duplex circuits -each character is transmitted independently -has a start and stop bit for each character
continuous ARQ
-sender does not wait for a NAK after sending a message, it immediately sends the next one
stop-and-wait ARQ
-sender stops and waits for a response from the receiver after each data packet, after receiving a packet, the receiver sends an ACK if the message is good -if packet is received with an error, then a NAK is sent
How does hamming code check for errors in the bits? and how does it fix these errors?
0 1 0 0 1 1 1 1 2 3 4 5 6 7 P1 = 0 + 0 + 1 + 1 = 0 P2 = 1 + 0 + 1 + 1 = 1 P4 = 0 + 1 + 1 + 1 = 1 Syndrome: 110 → error. There is a 1. 110 in binary = 6. That means position 6 is flipped. So it goes back and flips the 6th bit. if it equals 0, there is no error. If there is at least a 1, then there is an error.
Multicast MAC
01-00-5e (start)
Major Functions of the Data Link Layer are..
1) Media Access Control 2) Error Control 3) Message Delineation
2 Possible Approaches to Media Access Control (MAC): **Exam Q
1)Contention Based 2)Controlled Access ** Exam Q
What are the two categories of network errors?
1. Corrupted data 2. Lost data
Ideal MAP
1. When a single node wants to transmit it is able to use the entire available bandwidth, R 2. When more than one node, M nodes, wish to transmit they can transmit at an average rate of R/M 3. Fully decentralized -- No special transmission controller node -- No clock synchronization. 4. Simple
Two approaches to media access control
1. contention access 2. controlled access
Switch filtering algorithm
1. record frame destination and source 2. index table using mac destination address 3. if (entry found for destination){ if(entry destination same as source){ drop }else{ send to destination mentioned in frame }else{ flood network }
Modern advanced ethernet has gone up to
10 Gbps
Traditional ethernet
10 Mbps
Fast ethernet
100 Mbps
EtherType field
16 bits and used to describe the protocol of the contents of the frame.
How many types of addresses are typically defined by the Link Layer protocols?
3
How many bytes are IPv4 addresses?
4
Payload
46-1500 bytes, usually IP packet (sometimes control info)
MAC
48-bit binary value expressed as 12 hexadecimal digits
Ethernet (IEEE 802.3)
A LAN developed by the Xerox Corporation. It uses coaxial cable or twisted-pair wires to connect the stations. It was standardized as IEEE 802.3.
Hub and spoke
A WAN version of the star topology in which a central site interconnects branch sites using point-to-point links.
parity bit
A binary bit appended to an array of bits to make the number of 1 bits always be odd or even for an individual character. For example, odd parity may require three 1 bits and even parity may require four 1 bits.
stop bit
A bit that follows the group of bits representing a character. Used to signal the end of a character in asynchronous transmission.
start bit
A bit that precedes the group of bits representing a character. Used to signal the arrival of the character in asynchronous transmission.
high-level data link control (HDLC)
A bit-oriented protocol in which control of data links is specified by series of bits rather than by control characters (bytes).
Frame
A block of data
High-level data link control (HDLC)
A formal standard developed by the ISO often used in WANs
Hamming code
A forward error correction (FEC) technique named for its inventor.
Ethernet frame
A highly structured collection of info presented in a specific order.
What is one of example of a cause in intermodulation noise?
A maladjusted modem may transmit a strong frequency tone when not transmitting data, thus producing this type of noise.
contention
A method by which devices on the same shared multipoint circuit compete for time on the circuit.
token passing
A method of allocating network access wherein a terminal can send a message only after it has acquired the network's electronic token.
Automatic Repeat reQuest (ARQ)
A receiver that detects an error simply asks the sender to retransmit the message until it is received without error.
burst error
A series of consecutive errors in data transmission. Refers to the phenomenon on communication circuits in which errors are highly prone to occurring in groups or clusters.
Intermodulation noise
A special type of cross-talk. The signals from two circuits combine to form a new signal that falls into a frequency band reserved for another signal.
Automatic Repeat reQuest (ARQ)
A system employing an error-detecting code so conceived that any error initiates a repetition of the transmission of the incorrectly received message.
forward error correction
A technique that identifies errors at the received station and automatically corrects those errors without retransmitting the message.
There are two commonly used controlled access techniques. What is one of them?
Access requests
How do we detect the start of messages?
Add a special start code to the beginning of each frame. The start code is a unique pattern that only occurs at the start of each frame. It enables synchronisation after error - wait for the nest start code, then begin reading frame headers, etc...
What is the Parity error detection method?
Add one extra bit for every byte (or whatever you decide) so there is even number of 1's or 0's. Parity bit is 1 if we have odd number of 1's. Parity bit is 0 if we have even number of 1's.
What is the checksum error detection method?
Add up all the bytes in a message and send the sum. (TCP/UDP/IP uses this).
Modular Arithmetic
Adding numbers
What does ARP stand for?
Address Resolution Protocol
ARP
Address resolution protocol - Each node on a network has an arp table -- Arp tables map MAC addresses to IP addresses -- IP address; MAC address; TTL - TTL(Time to live) is the time that the mapping will remain in the table before deletion -- A mappings TTL will be renewed when a connection to the node holding the table is made from that attribute -- 20 minutes by default - network layer
Bus
All end systems are chained to each other and terminated in some form on each end. Infrastructure devices such as switches are not required to interconnect the end devices.
Broadcast
All hosts that receive the frame (all 1s in binary or all Fs in hexa)
Synchronous transmission
All the letters or data in one group of data are transmitted at one time as a block of data
Suppose that we are using bit stuffing, and the flag byte that is found at the beginning and at the end of each frame is 00000000 (notice that this is different from the scheme adopted in HDLC). Design a bit stuffing scheme that allows for unambiguous recognition of frames. Given the data frame 00110110 00000000 00001101 (flag bytes excluded), give the stuffed frame according to your scheme. (7 marks) 2014
All the required scheme has to do is to avoid sequences of more than seven 0s in the payload. Hence, we stuff a 1 after every sequence of seven 0s. The given frame after the stuffing is therefore : 00110110 000000100 00001101
How can we correct errors?
Also send error correction codes in each frame, allows receiver to correct (some) errors without having to contact the sender.
cyclical redundancy check (CRC)
An error-checking control technique using a specific binary prime divisor that results in a unique remainder. It usually is a 16-to 32-bit character.
What is internet checksum?
Another error detection method. Sum the data values and send checksum in each frame. The receiver recalculates, if there's a mismatch, bit error. This is better than parity codes because it detects multiple bit errors.
attenuation
As a signal travels through a circuit, it gradually attenuates, or loses power. Expressed in decibels, attenuation is the difference between the transmitted and received power caused by loss of signal strength through the equipment, communication circuits, or other devices.
Contention Based
Computer wait until circut is free and then tranmit. ; first come first serve basis
Contention
Computers wait until the circuit is free and then transmit whenever they have data to send. Commonly used in Ethernet LANs.
Vendor
Assigned by vendor, last 3 bytes
Hub
Common communication point for devices
Repeater
Commonly used on digital circuits. Receives the incoming signal, translates it into a digital message, and retransmits the message.
Node
Communication endpoint (connection point or redistribution point)
polling
Any procedure that sequentially queries several terminals in a network.
What does ARQ stand for?
Automatic Repeat Request
Length
Bigger than 0x0600
full-duplex communication
Both devices can transmit and receive on the media at the same time.
half-duplex communication
Both devices can transmit and receive on the media but cannot do so simultaneously.
Continuous ARQ (full-duplex)
Both sides agree on the size of the sliding window, sender contnues sending packets without waiting for the reciever to ACK, reciever continues recieving messages without ACKing them right away, once secnder has transmitted, the max number of packets permitted in the sliding window, it can send more packets until the reciever sends the ack
Network topology types
Bus, star, ring, full mesh
collision avodiance
CMSA/CA
collision detection
CMSA/CD
CSMA
Carrier sense multiple access - Wait for line to be idle then transmit - If multiple nodes realize that the channel is idol then they can still transmit simultaneously - When a collision occurs entire packet delivery time is wasted
CSMA/CD
Carrier sense multiple access w/ collision detection - Same as CSMA but more - Collision detection -- easy in wired lan can detect signal strength -- harder in wireless lan since transmissions further away can be masked by closer signals - Binary exponential back off -- every time a collision occurs double the time till re-transmit. - efficiency = 1/(1+(5*Tprop)/Ttrans) - Better performance than ALOHA
White noise/ Gaussian noise
Caused by the thermal agitation of electrons and therefore is inescapable.
What is encoding NRZI?
Change level for a 1. Unchanged level for a 0.
Example of an Even Parity
Character: 'A' Binary: 01000001 Parity Bit: 0 Character: 'C' Binary: 01000011 Parity Bit: 1
Star
Each node connects to a central device (common)
Access Request Technique
Client computers that want to transmit send a request to transmit to the device that is controlling the circuit. Controlling device grants permission for one computer at a time to transmit
What if the client doesnt have data?
Client responds negatively and server or controlling device asks the next client
Full duplex (switch)
Collision free links to hosts
CSMA/CD process
Connects to shared medium, sender listens until medium is not busy, if collision occurs it waits and resents after a random amount of time
Trailer
Contains control information added to the end of the PDU
CAM
Content addressable memory
What is one of the fundamental approaches to media access control?
Contention
Which MAC approach is better suited for small networks that have low usage?
Contention is better suited because there is little chance of collision.
What is one of the fundamental approaches to media access control?
Controlled Access
Which MAC approach is better suited for large networks that have high usage?
Controlled is better suited because it prevents collisions and makes more efficient use of the circuit.
Roll-call Polling
Controller works consecutively through a list of clients, first polling client 1, then client 2, and so on, until all are polled
A ________ devics periodically polls the client if it has data
Controlling
What is the MAC responsible for at the sending computer?
Controls how and when the physical layer converts bits into the physical symbols that are sent down the circuit.
Controlled access
Controls the circuit and determines which clients can transmit at what time.
Fragment-free
Cut-through variation, receives first 64 bytes, checks and forwards
CRC
Cyclic redundancy check - View data as a binary number, D - a generator is chosen heuristically, G -- Generator must be r+1 bits long -- Both the sender and receiver know what G is - Divide D*2^r by G -- Calculating R requires you to multiply D by 2^r - Can detect all burst errors less than r+1 length.
Error detection/correction
Cyclic redundancy check (CRC) Data is divided by a generator known to the sender and receiver. Parity Data inside of the frame is cut up into sections of size d, at the end of every section a single bit is added. - Even parity - Odd parity - 2D parity
Which layer creates the frame PDU?
Data Link Layer
What are the sublayers of Data Link Layer?
Data link control and media access control
Bus
Each node connects to a common medium
encoding
Dependent on the media. Putting the data into the correct language for the specified media
Cancellation
Designers now pair wires in a circuit. When two wires in an electrical circuit are placed close together, their magnetic fields are the exact opposite of each other. Therefore, the two magnetic fields cancel each other and also cancel out any outside EMI and RFI signals.
Error Control
Detecting and correcting transmission errors
What is framing?
Determines start and end of packets and their order by inserting packet headers. Sending data in a specific format. The problem is to know the sequence of when the frame starts and ends.
Data Link Protocol
Determines who can transmit at what time, where a message begins and ends, and how a receiver recognizes and corrects a transmission error.
controlled access
Each node has its own time to use the medium. These deterministic types of networks are inefficient because a device must wait its turn to access the medium. Legacy token ring LANs are an example of this type of access control.
Ring
Each node has two connections (data is in one end and out the other)
Which type of noise in fiber-optic cables is almost always caused by poor connections?
Echo. The solution is to tune the transmission equipment and redo the connections.
Responsibilities:
Encoding, Sending/Receiving and framing data, error checking and recovery/retransmission
star
End devices are connected to a central intermediate device
Ring
End systems are connected to their respective neighbor
Flow control
Ensuring that the computer sending the message is not transmitting too quickly for the receiver
Store-and-forward
Entire frame is received and stored in memory, error checked and forwarded
What are services the Data Link Layers do not provide?
Error Congestion
Contention based controll is commonly used in which type of LANs?
Ethernet
Lan protocol
Ethernet
What does 4B/5B encoding do?
Every 4 bits of data are encoded in 5 bits of signal. Example: • 0000 → 11110 • 1111 → 11101 • 0001 → 01001 • 1110 → 11100
1. The user data is segmented by the transport layer, placed into packets by the network layer, and further encapsulated into frames by the data link layer 2. The physical layer encodes the frames and creates the electrical, optical, or radio wave signals that represent the bits in each frame 3. these signals are then sent on the media, one at a time 4. The destination node physical layer retrieves these individual signals from the media, restores them to their bit representations, and passes the bits up to the data link layer as a complete frame
Explain each step of the process that data undergoes from a source node to a destination node (hint there are 4 steps)
ARPs job is to query the local network with a MAC address and find its IP address. t/f
False
All ethernet ports in a router use the same MAC address t/f
False
Ethernet ports can only have one IP address t/f
False
Routers can only have one IP and Link Layer address t/f
False
The source and destination link addresses and the IP address change with each hop. t/f
False
The Data Link uses IP address to deliver frames. t/f
False.
Form factors
Fixed, modular, stackable
synchronous transmission
Form of transmission in which data is sent as a fixed-length block or frame. Compare with asynchonous transmission.
Data link
Forwards frames, uses MAC address
PDU
Frame
Describe the problem of framing in the Data Link layer. Describe the byte count technique for framing, and explain why it is not a viable option to address the framing problem. (7 marks) 2014
Framing means breaking up the data stream into chunks (frames) so that the receiver can use checksum techniques to detect errors frame by frame. The byte count technique consists of starting each frame with a byte containing the length of the frame (represented as an integer) in bytes. This technique is particularly error-prone as any error occurring in the first byte of a frame will incorrectly break the subsequent frame boundaries at the receiver, propagating errors.
Describe the problem of framing in the Data Link layer. Describe the byte count technique for framing, and explain by means of an example why it is not a viable option to address the framing problem. (7 marks)
Framing means breaking up the data stream into parts (frames) so that the receiver can isolate each frame and provide its services. In particular using checksum techniques to detect errors frame by frame. The byte count technique consists of starting each frame with a byte containing the length of the frame (represent as integer) in bytes. This technique is particularly error-prone as ane error occurring in the first byte of a frame will incorrectly break the subsequent frame boundaries at the receiver, propagating errors. The STUDENT should show an example where the receiver interprets as the first byte of a frame a byte that is in fact a payload byte. Subsequent frame are therefore highly unlikely to be correctly identified (unless a rare coincidence happens)
FDMA
Frequency division multiple access - Channel partitioning - channel divided into multiple frequency bands - unused transmission bands idle - not time sensitive - each node uses its own frequency band
Ethernet switches
Full duplex, form-factors
frame
Generally, a group of data bits having bits at each end to indicate the beginning and end of the frame. Frames also contain source addresses, destination addresses, frame type identifiers, and a data message.
Assume we have an error correcting code made of four code words: 0000000000, 0000011111, 1111100000 and 1111111111. Define the Hamming distance between two sequences of bits and determine the minimum Hamming distance between any two words in the code above. Determine how many 1-bit errors (within a code word) can be corrected by this code, and how many 1-bit errors (within a code word) can be detected. Explain how correction and detection work in this case. Determine the code rate of this code, defined as follows: if m bits are sufficient for encoding all the words in the code but we are using n bits per word, the code rate is m/n. (10 marks)
Given two bit sequence of the same length, the Hamming distance is the number of bits that need to be changed in order to obtain the first sequence from the second (or vice-versa). It is obtained by XORing the two sequences and counting the number of ones in the result. The minimum mutual Hamming distance among any pair of the given code words is d = 5. An error is detected by checking if the received words are actually coded words. As for correction, if a received word is not a codeword, it is interpreted as the closest codeword in terms of Hamming instance. Since to detect n errors we need a Hamming distance of n + 1, with our code we detect up to 4 errors (within the same codeword). Since to correct n errors we need a Hamming distance of 2n + 1, our code corrects up to 2 errors (within the same codeword). Finally, since 2 bits are sufficient to encode 4 codewords and we are using 10 bits per codeword, the code rate is 1/5 (20%).
Assume we have an error correcting code made of four codewords: 000000, 000111, 111000 and 111111. Define the Hamming distance between two sequences of bits. State all six Hamming distances between pairs of the above codewords. Determine how many 1-bit errors (within a codeword) can be corrected by this code, and how many 1-bit errors can be detected. Explain how correction and detection work in this case. (10 marks) 2012
Given two bit sequences of the same length, the Hamming distance is the number of bits that need to be changed in order to obtain the first sequence from the second (or vice-versa). It is obtained by XORing the two sequences and counting the number of ones in the result. The mutual Hamming distance among any of the given two codewords is d = 3. An error is detected by checking if the received words are actually codewords. As for correction, if a received word is not a codeword, it is interpreted as the closest codeword in terms of Hamming distance. Since to detect n errors we need a Hamming distance of n + 1, with our code we can detect 2 errors. Since to correct n errors, we need a Hamming distance of 2n + 1, our code corrects single errors.
What is the Hamming Code?
Gives a method for constructing a code with distance k=3. Uses n = 2^k - k - 1 = 2^3 - 3 - 1 = 4
What addresses do data link protocols like Ethernet use?
Global scope addresses.
Host addresses may be link-local or...?
Global scope.
Multicast
Group of hosts
Give an example of an error correcting code.
Hamming code.
HDLC frame breakdown
Header (flag, DA, control, type), payload, trailer (FCS)
Frame breakdown
Header, payload, trailer
What is Phase Shift Keying?
High for 1's, low for 0's.
HDLC
High-level data link control, for WAN, only for point to point links and synchronous links
What is Reliable Transmission?
How do we send a packet reliably if we know that it can be lost somewhere in the network.
Used to provide always on broadband services to homes and small buisnesses
How is fiber-optic cabling used in Fiber-to-the-home (FTTH)
It is used for backbone cabling applications and interconnecting infrastructure devices
How is fiber-optic cabling used in enterprise networks?
used by service providers to connect cities and countries
How is fiber-optic cabling used in long-haul networks?
used to provide reliable high-speed, high-capacity solutions capable of surviving in harsh undersea environments up to transoceanic distances
How is fiber-optic cabling used in submarine cable networks?
Media Sharing
How the nodes share the media
LLC is defined by
IEEE 802.2
Ethernet
IEEE 802.3
Type/Length
Identifies either type of network or length of frame
Collision
If a node receives two or more signals at the same time. Data from both transmissions interfere and information is corrupted
Data Link Layer provides service to the __
Network Layer
Briefly describe the main ideas behind sliding windows protocols. Explain the notion of piggybacking in such protocols, and why piggybacking is useful. (11 marks) 2014
In sliding windows protocols, two parties transmit in both directions. Frames are acknowledged by sending ack messages. The attentive student will point out that acknowledgements are sent in a delayed fashion - that is, acknowledgements are attached to the next data frame that is transmitted; this technique is known as piggybacking. Frames are numbered from 0 up to some maximum. Sender and transmitted keep a sending and a receiving window, respectively; each window is an interval of frame sequence numbers. At the sender, each packet to be sent is given the next sequence number outside the window, and the upper edge of the window is advanced by one; whenever an acknowledgement (for the packet having the lowest sequence number in the window) is received, the lower edge of the window is advanced by one. The sending window serves as a buffer for unacknowledged packets that may need to be retransmitted. At the receiver, any frame falling within the window is accepted and stored; if the frame with the lowest sequence number in the window arrives, both the lower and the upper edge of the window are advanced by one. Piggybacking is a technique of sending ack signals in a delayed (or lazy) fashion; instead of being send by
IEEE
Institute of Electrical and Electronics Engineers.
What is Frequency Shift Keying?
It depends on how fast (close) the frequencies (waves) are to each other. Depends on the distance between the middle of the frequencies.
Why is NRZ a bad idea?
It is hard to track the number of 0's and 1's if there are multiple of the same bits next to each other. Every transition "makes a noise" and that is all you hear.
What makes a good start code?
It must not appear in the frame headers, data or error detecting code. It must allow timing recovery (as stated above...).
What is Amplitude Shift Keying?
It will send waves with differing amplitudes to indicate 1's and 0's. Waves with a higher amplitude represent 1's. Waves with lower amplitudes (or no amplitude) represent 0's
What happens when the frame is lost in reliable transmission?
It would send the frame again.
What is the LLC responsible for at the sending computer?
It's responsible for taking the network layer PDU- usually an IP packet- and surrounding it with a data link layer PDU- often an Ethernet frame.
What is the LLC responsible for at the receiving computer?
LLC sublayer software removes the data link layer PDU and passes the message it contains (usually an IP packet) to the network layer software.
LLC
Logical link layer
Data Link
Layer 2 in the Internet model
What is Source of Errors
Line Noise and Distortion; Undesirable electrical signal, introduced by equipment and natural Disturbances; missing bits, extra bits, flipped bits
Switch MAC address table
Lists of each learned MAC address and its corresponding port, may use CAM
collision
M.A.C. - contention - when devices transmit at the same time, a ______ occurs
Every device (worldwide) has a unique
MAC address
Sublayers
MAC and LLC
media access control, error control, message dilineation
Major functions of a data link layer protocol
Ethernet
Many speeds from 10 Mbps to 40 Gbps
MAC
Media Access Control - Unique to every host - Companies that produce devices capable of net communication buy MAC address ranges from IEEE. - 48 bit -- split up into 6 8bit chunks, consisting of 2 Hexadecimal numbers(4bit words) - Flat. The same everywhere.
MAC Address
Media Access Control address. globally unique identifier attached to an individual network interface. a way to identify which node (computer) the transmission was meant for. a 48 bit number normally represented by six groupings of 2 hexadecimal #'s
Burst error
More than 1 bit is changed by the error-causing condition
MPLS
Multi protocol label switching - avoids complex lookups in routing tables -- uses fixed length identifiers as opposed to shortest prefix - keeps ip address - implements traffic engineering -- Has precalculated backup routes - MPLS routing decisions can differ from IP - MPLS is protocol independent and allows the use of multiple network layer protocols in a network.
Physical links can be point-to-point or...?
Multi-access.
Error Control
Network Errors - Types - Corrupted Data - Lost Data - Caused by problems in transmission (not humans) Networks should be designed with: - Error Prevention - Error Detection - Error Correction
NIC
Network interface card - Outside of CPU - Contains hardware and firmware - Consists of a controller attached to a physical wire link
Did the previous methods correct errors?
No, only detected them.
Which nodes will decapsulate frames in the Data Link Layer?
Nodes in between source and destination Destination Node
What are the terms used for hosts/routers and the networks in between the in the Data Link Layer?
Nodes, Links
Why do we need error detection?
Noise an interference at physical layer can cause bit errors, more common in wireless systems.
impulse noise
Noise caused by individual impulses on the circuit.
Gaussian noise
Noise that is characterized statistically by a Gaussian, or random, distribution.
What does NRZ stand for?
Non-Return to Zero
What NRZI stand for?
Non-Return to Zero Inverted
Asynchronous transmission
Often referred to as start-stop transmission because the transmitting computer can transmit a character whenever it is convenient, and the receiving computer will accept that character.
Point-to-Point Protocol
Often used in WANs. Designed to transfer data over a point-to-point circuit but provides an address so that it can be used on multiple circuits.
Within polling where do all the messages that need to be tranmitted get stored?
On the client side; if it has data it sends data
Hub polling (token passing)
One device starts the poll and passes it to the next computer on the multipoint circuit, which sends its message and passes the poll to the next.
OUI
Organizationally Unique Idnetifier. First 3 octetsof a MAC address. Assigned to individual hardware manufacturers by the IEEE
OUI
Organizationally unique identifier assigned by IEEE, first 3 bytes
Data Link
Organizes physical layers bit streams into coherent messages for the network layer
Point-to-Point Protocol (PPP)
PPP (multilink Point-to-Point Protocol) is an inverse multiplexing protocol for combining circuits of different speeds (e.g., a 64,000-bps circuit with a 14,400-bps circuit), with data allocated to each circuit is based on speed and need. PPP enables the user to change the circuits allocated to the PPP multiplexed circuit in mid-transmission so that the PPP circuit can increase or decrease the capacity. PPP is the successor to SLIP.
What are the three types of error detection methods?
Parity, checksum, and CRC (Cyclic Redundancy Check)
There are two commonly used controlled access techniques. What is one of them?
Polling
roll-call polling
Polling accomplished from a prespecified list in a fixed sequence, with polling restarted when the list is completed.
VLAN
Port based - Acts as if it were multiple physical devices - communications between two VLANS on the same switch must still be done through a router or switch -- even though they're on the same device - traffic isolation -- frames destined for one host on a VLAN can not be sent to the other VLAN - Trunking -- One of the ports in a vlan can lead to another physical switch which contains further isolated links for this vlan -- Allows the spanning of multiple switches
Ethernet frame breakdown
Preamble (before), header (destination MAC, source MAC, T/L), payload, trailer (FCS)
Impulse noise (spikes)
Primary source of errors in data communications. It's heard as a click or crackling voice and can last as long as 1/100 of a second.
Shielding
Protecting wires by covering them with an insulating coating. One of the best ways to prevent impulse noise, cross-talk, and intermodulation noise.
What is a PDU?
Protocol Data Unit
Cut-through
Receives destination MAC address, starts forwarding frame while still receiving kt
Media Access Control
Refers to the need to control when computers transmit. With point-to-point full-duplex configurations, MAC is unnecessary because there are only two computer on the circuit, and full duplex permits either computer to transmit at any time.
Physical topology
Refers to the physical connections and identifies how end devices and infrastructure devices such as routers, switches, and wireless access points are interconnected.
Logical topology
Refers to the way a network transfers frames from one node to the next. This arrangement consists of virtual connections between the nodes of a network
Data Link Layer
Responsible for moving a message from one computer or network device to the next computer or network device in the overall path from sender to receiver
Data Link
Responsible for moving messages from one device to another
Data Link Layer
Responsible for moving messages from one device to another; controls the way messages are sent on media; also organizes the physical layer bit streams into coherent messages for the network layer
CSMA/CA process
Same, bus can't listen for collisions so takes extra steps to avoid them
Continuous ARQ
Sender does not wait for an acknowledgement after sending a message, it immediately sends the next one.
What is the CRC error detection method?
Sender: 1) Extend the n (number of bits we're sending) data with k zeros 2) Divide by a generator value C 3) Use the remainder as the k check bits Receiver: 1) Divide and check for zero remainder
What is a SONET?
Synchronous Optical Networking
What is the preferred name for the PDU in Transport Layer of the TCP/IP model?
TCP segment
Throughput -Compute TRIB
TRIB = (K(M-C)(1-P)) / ((M/R) + T), where K = information bits per character M = frame length in characters R = data transmission rate in characters/sec C = average number of non-information characters per frame (control characters) P = probability that a frame will require retransmission T = time delay between frames (in seconds)
What is the MAC responsible for at the receiving computer?
Takes the data link layer PDU from the LLC sublayer, converts it into a stream of bits, and controls when the physical layer actually transmits the bits over the circuit. It also receives a stream of bits from the physical layer and translates it into a coherent PDU, ensures that no errors have occurred in transmission, and passes the data link layer PDU to the LLC sublayer.
Amplifier
Takes the incoming signal, increases its strength, and retransmits it on the next section of the circuit.
Data Link Layer
The Data Link Layer, also known as layer 2, is responsible for moving a message from one computer or network device to the next computer or network device in the overall path from sender or receiver. It controls the way messages are sent on the physical media. Both the sender and receiver have to agree on the rules, or protocols, that govern how they will communicate with each other. A data link protocol determines who can transmit at what time, where a message begins and ends, and how a receiver recognizes and corrects a transmission error.
Explain how the notion of Hamming distance is used in error correcting and error detecting codes. Assuming we want to detect d one-bit errors, what is the minimum distance required between codewords? Assuming we want to correct d one-bit errors, what is the minimum distance required between codewords? (11 marks) 2013
The Hamming distance two words (of the same length) is the number of bits in which the two words differ. Codes do not use all possible codewords, instead, codewords are chosen so that any pair of them has at least a Hamming distance h between them. Received words that are not codewords can be detected, and the closest (in term of Hamming distance) codeword. To detect d 1-bit errors we need h to be at least d + 1, while to correct d 1- bit errors we need h to be at least 2d + 1.
Collision
The attempt to talk simultaneously
flow control
The capability of the network nodes to manage buffering schemes that allow devices of different data transmission speeds to communicate with each other.
Describe the technique of bit stuffing in the HDLC (High-level Data Link Control) protocol. Suppose we want to send the following sequence of bits: 0011111101110111111111001. Give the transmitted sequence after bit stuffing. Hint: remember that bit stuff- ing ensures that the flag byte 01111110 (which delimits frames) need not appear in the payload. (7 marks) 2013
The flag byte 01111110 delimits the start and the end of frames. To avoid the transmission of flag bytes in the payload, a 0 is added after every sequence of 5 consecutive ones. The sequence will therefore be transmitted (excluding the flag) as 0011111¯010111011111¯01111001 where the stuffed bits are marked with a bar.
Early timeout in reliable transmission? ACK sent after the timeout, it is late.
The frame will be sent again, and a new ACK will be sent. That frame that was already sent is sent again and acknowledged as a new frame.
What happens when the ACK is lost in reliable transmission?
The frame would be sent again. The frame that was already sent is sent again and acknowledged as a new frame.
Assume we are using parity bits with interleaving, with n = 8 columns and k = 3 rows. If we have to transmit the three-byte sequence 01011100 11000010 00111100 (made of 24 data bits), how are the parity bits added to the sequence? What is the longest error burst that the resulting code is able to detect? Justify your answer.
The k words are arranged in columns, and the added parity word makes the sum (modulo 2) of the bits in each column equals to 0. 01011100 11000010 00111100 10100010 An error burst of n + 1 bits will affect two bits in the same column (the initial and final bits, which are error bits by definition of burst), and could therefore go undetected (assuming there is at most one in each kn-bit sequence)
Assume we are using parity bits with interleaving, with n = 8 columns and k = 4 rows. If we have to transmit the four-byte sequence 00011101 01010101 11001001 01101100 (made of 32 data bits), how are the parity bits added to the sequence? What is the longest error burst that the resulting code is able to detect? Justify your answer. (8 marks)
The k words are arranged in columns, and the added parity word makes the sum (modulo 2) of the bits in each column to 0. 00011101 01010101 11001001 10101101 - parity bits An error burst of n + 1 bits will affect two bits in the same column (the initial and final bits, which are error bits by definition of burst), and could therefore go undetected. On the contrary, all bursts of length at most n will be detected (assuming there is at most one in each k-bit sequence)
logical link control (LLC) sublayer
The logical link control, or LLC, sublayer is just an interface between the MAC sublayer and software in layer 3 (the network layer) that enables the software and hardware in the MAC sublayer to be separated from the logical functions in the LLC sublayer. By separating the LLC sublayer from the MAC sublayer, it is simpler to change the MAC hardware and software without affecting the software in layer 3. The most commonly used LLC protocol is IEEE 802.2.
Acknowledgement (ACK)
The packet was received without error
Why is framing and synchronisation required?
The physical layer provides an unreliable bit stream. - Bits may be corrupted. - Timing can be disrupted.
What does NRZI solve that NRZ does not solve?
The problem is solved for the 1's because it transitions between levels every time there is a 1. But, if there are multiple 0's, then it is hard to count them.
Polling technique
The process of sending a signal to a client computer that gives it permission to transmit. Clients store all messages that need to be transmitted.
Echoes
The result of poor connections that cause the signal to reflect back to the transmitting equipment.
negative acknowledgment (NAK)
The return signal that reports an error in the message received. The opposite of acknowledgment (ACK).
Stop-and-wait ARQ
The sender stops and waits for a response from the receiver after each data packet.
Suppose A and B are respectively a sender and receiver station using a stopand-wait sliding window protocol (with window size 1), and that: • A has to send the sequence of packets A0, A1; • B has to send the sequence of packets B0, B1. Describe the sequence of messages sent during the transmission, using the notation h sequence no, acn no, packet i for a message. Assume that A initiates the communication, and B starts transmitting only after having received the first message from A. Also assume that no messages are lost and that there are no transmission errors. (11 marks) 2015
The sequence is a normal one for the stop-and-wait, where no messages are lost or discarded. A sends (0, 1, A0) B gets (0, 1, A0) and accepts it B sends (0, 0, B0) A gets (0, 0, B0) and accepts it A sends (1, 0, A1) B gets (1, 0, A1) and accepts it B sends (1, 1, B1) A gets (1, 1, B1) and accepts it
Collision domain
The shared communication medium for multiple devices (hub)
throughput
The total amount of useful information that is processed or communicated during a specific time period.
Transmission Efficiency
The total number of information bits divided by the total bits in transmission.
Throughput
The total number of information bits received per second, after taking into account the overhead bits and the need to retransmit frames containing errors.
Suppose we are using, as error detecting code, a single parity bit (in even parity) for each transmitted (8-bit) byte, and that we want to send the sequence 1010001 0001001. Explain how the parity bits are added. How many 1-bit errors can such a code detect? (7 marks) 2013
The transmitted sequence consists of the following two bytes: 10100011 00010010 Such code is "blind" to any even number of 1- bit errors (in the same byte). It is therefore guaranteed to detect only single errors.
point-to-point
This is the simplest topology that consists of a permanent link between two endpoints. For this reason, this is a very popular WAN topology.
Media Access Control (MAC)
This lower sublayer defines the media access processes performed by the hardware. It provides data link layer addressing and access to various network technologies.
Mesh
This topology provides high availability, but requires that every end system be interconnected to every other system. Therefore, the administrative and physical costs can be significant. Each link is essentially a point-to-point link to the other node.
Logical Link Control (LLC)
This upper sublayer communicates with the network layer. It places information in the frame that identifies which network layer protocol is being used for the frame. This information allows multiple layer 3 protocols, such as IPv4 and IPv6, to utilize the same network interface and media.
copper, fiber-optic cables, wireless
Three basic forms of network media
TDMA
Time division multiple access - Channel partitioning - Divides timeframe into slots which repeats -- slot length = transmission time - nodes connected to this node will have designated slots to transmit - unused slots go idle
What is the purpose of the data link layer?
To arbitrate access to the physical layer: - Identify devices (addressing). - Structure and frame raw bit stream - Detect and correct bit errors. - Control access to channel (media access control). Overall, turns raw bit stream into a structured communications channel.
FCS
Trailer, frame check sequence (that frame sent is identical to the received), uses 32 but CRC (cyclic redundancy check)
asynchronous transmission
Transmission in which each information character is individually synchronized, usually by start and stop bits. The gap between each character is not a fixed length. Compare with synchronous transmission.
Which layer creates the segment PDU?
Transport layer
Broadcast address points to all points within the same network. t/f
True
How many sublayers is the Data Link Layer divided into?
Two
Overhead bits
Used for purposes such as error checking and marking the start and end of characters and packets.
Information bits
Used to convey the user's meaning
Link Access Protocol for Modems [LAP-M]
Used when retransmitting only the packets which contained an error
Go-Back-N ARQ
Used when retransmitting the first packet with an error and all those that follow it.
Forward error correction
Uses codes containing sufficient redundancy to prevent errors by detecting and correcting them at the receiving end without retransmission of the original message.
What are the primary sources of impulse noise?
Voltage changes in adjacent lines, lighting flashes during thunderstorms, fluorescent lights, and poor connections in circuits.
1.Allowing the upper layers to access the media 2. Accepting layer 3 packets and packaging them into frames 3. preparing network data for the physical network 4. Controlling how data is placed and received on the media 5. Exchanging frames between nodes over a physical network media, such as UTP or fiber-optic 6. Receiving and directing packets to an upper layer protocol 7. performing error detection
What is the data link layer of the OSI model responsible for? (hint 7)
Shared medium
WLANs operate in half duplex, which means only one device can send or receive at a time. The wireless medium is shared amongst all wireless users. The more users needing to access to WLAN simultaneously, results in less bandwidth for each user.
What if the start code appears in the data?
We use bit stuffing. The sender inserts a 0 bit after sending any five consecutive 1 bits - unless sending the start code. When the receiver gets the data, if it sees five consecutive 1 bits, it looks at the sixth bit: - if 0, has been stuffed so remove - if 1, look at seventh bit: - if 0, start code - if 1, corrupt frame
Unshielded twisted pair (UTP) Shielded twisted pair (STP) coaxial
What are the three main types of copper media used in networking?
Source of Error: Attenuation
What causes it: Gradual decrease in signal over distance How to prevent or fix: Use repeaters
Source of Error: White Noise
What causes it: Movement of electrons How to prevent or fix: Increase signal strength
Source of Error: Cross-talk
What causes it: Multiplexer guard bands too small or wires too close together How to prevent or fix: Increase the guard bands or move or shield the wires
Source of Error: Echo
What causes it: Poor (misaligned) connections How to prevent or fix: Fix the connections or tune equipment
Source of Error: Intermodulation Noise
What causes it: Signals from several circuits combine How to prevent or fix: Move or shield the wires
Source of Error: Impulse Noise
What causes it: Sudden increases in electricity (e.g. lightning) How to prevent or fix: Shield or move the wires
1. cable length 2. order wire map 3. crosstalk 4. signal loss due to attenuation
What should a UTP cable tester be able to test for? (hint 4)
1. Wireless access point 2. Wireless NIC adaptor
What two devices are required in order for a wireless network to function?
When does a collision take place in Contention Based Access Control?
When 1+ computers transmit at the same time; after collision, you need to determine which computer is allowed to transmit first
Cross-talk
When one circuit picks up signals in another. Increases with increased communication distance, increased proximity of the two wires, increased signal strength, and higher-frequency signals
intermodulation noise
When signals from two independent lines intermodulate. A new signal forms and falls into a frequency band differing from those of both inputs. The new signal may fall into a frequency band reserved for another signal.
CSMA/CD
Wired, carrier sense multiple access / collision detect
Controlled Access is commonly used with which type of networking?
Wireless Lans and Mainframes
Security
Wireless communication coverage requires no access to a physical strand of media. Therefore, devices and users not authorized for access to the network, can gain access to the transmission.
coverage areas
Wireless data communication technologies work well in open environments. However certain construction materials used in buildings and structures, and the local terrain, will limit the effective coverage
interference
Wireless is susceptible to interference and can be disrupted by such common devices as household cordless phones, some type of fluorescent lights, microwave ovens, and other wireless communications.
CSMA/CA
Wireless, carrier sense multiple access / collision avoidance
Describe the technique of byte stuffing for framing. Assume that we are to transmit the following sequence of bytes: A, B, FLAG, ESC, FLAG, C. Write the byte sequence after byte stuffing. (8 marks) 2015
With byte stuffing, the end of the frame is marked by the FLAG byte. If a FLAG byte is in the payload, it is preceded by an added ESC byte so that it is not interpreted as a FLAG by the receiver. The same is done if an ESC byte is in the payload. After stuffing the sequence in question is A, B, ESC, FLAG, ESC, ESC, ESC, FLAG, C.
Describe the technique of byte stuffing for frame detection. Assume we want to transmit the following sequence of five bytes: A, FLAG, FLAG, B, ESC (FLAG and ESC are the usual flag and escape byte, respectively). What is the byte sequence after stuffing? (6 marks) 2012
With byte stuffing, the end of the frame is marked by the FLAG byte. If a FLAG byte is in the payload, it is preceded by an added ESC byte. In turn, if ESC byte us in the payload, it is also preceded by an added ESC byte. After stuffing the sequence in question is A, ESC, FLAG, ESC, FLAG, B, ESC, ESC
Can error be caused by human error?
Yes - a mistake in anumber or by an application program
Can wired links be multi-access?
Yes, several hosts can be connected to a single cable to form multi-accesses.
Are there any other means of correcting errors other than error correcting codes?
Yes, upon detecting an error, we can simply request retransmission.
crossover
a cable used to interconnect similar devices (router to a computer uses a crossover because although they do not look the same they have the same function)
a multimode fiber cable
a orange jacket indicates....
Automatic Repeat reQuest (ARQ)
a receiver that detects an error simply asks the sender to retransmit the message until it is received without error
unicast
a transmission always meant for just one receiving address.
Ethernet (IEEE 802.3)
a very popular LAN protocol conceived by Bob Metcalfe and developed jointly by Digital, Intel, and Xerox in the 1970s --> it has been refined and developed into this standard that uses a contention media access protocol
a single mode fiber cable
a yellow jacket indicates...
Mash
all devices are connected
extended star
additional ethernet switches interconnect other star topologies
parity check
adds a specific number to the end of the frame depending on even or odd to check -if two bits are switched then parity check will not detect any error -easy to see the parity can detect errors only when an odd # have been switched; any even # of errors cancel one another out --> probability of working is 50%
overhead bits
bits used for purposes such as error checking or making the start and end of character packets (ex- parity bits)
information bits
bits used to convey the user's message
contention based access
all nodes operating in half-duplex compete for the use of the medium, but only one device can send at a time. However, there is a process if more than one device transmits at the same time. Ethernet LANs using hubs and WLANs are examples of this type of access control.
Switches
allow for the segmentation of the LAN into separate collision domains
Data Packet
an all encompassing term that represents any single set of binary data being sent across a network link, a concept, one set of data being sent from point A to point B
What is the command to get the IP address and MAC address that are reporting on your LAN?
arp -a
Hamming code
associates even parity bits with unique combinations of data bits
CRC Cyclical Redundancy Check
basically a mathematical equation that checks to make sure the same amount of data sent is received.
error rate
bit in error for every bit transmitted -the average error rate is 1/500,000
Data Link
controls the way messages are sent on media
efficiency
capacity used to transmit error detecting data compared to capacity to transmit the message itself
CSMA/CD
carrier sense multiple access with collision detection. used to determine when the communications channels are clear and when the device is free to transmit data.
bandwidth
carrying capacity of the medium
line noise
caused by equipment and natural disturbances
impulse noise
caused by sudden increase electricity -heard as a click or crackling noise and can last as long as 1/100 of a second -can be prevented by shielding or moving wires
white noise/Gaussian noise
caused by thermal agitation of electrons and inescapable -familiar as background hiss or static on radios and telephone -can be prevented by increasing signal strength
Ring
circular connection with tokens
access request
client computers that want to transmit send a request to transmit to the device that is controlling the circuit
Access Request
clients first send a request to device controlling the circuit and the control device with grant permission for one device at a time to transmit. When one computer has permission to transmit, all other computer wait until that computer has permission. When they have something to transmit they use a contention technique to send an access request
contention
computers wait until the circuits is free and then transmit whenever they have data to send --> commonly used with Ethernet & LANs
Wireless access point (AP)
concentrates the wireless signals from users and connects to the existing copper-based network infrastructure, such as ethernet. Home and small business wireless routers integrate the functions of a router, switch, and access point into one device
multimode fiber (MMF)
consists of a larger core and uses LED emitters to send light pulses
Single mode fiber (SMF)
consists of a very small core and uses expensive laser technology to send a single ray of light. Popular in long distance situations spanning hundreds of kilometers
Header
contains control information such as addressing
data
contains the IP header, transport layer PDU, and data
Roll Call Polling
controller works consecutively through a list of circuits and checks each client to see if they want to transmit, involves waiting because controller has to poll clients and wait for a response - Negative Response (no data to be sent) -Needs a time to prevent lock-up - A, B, C, D, E , .... A, B, C, D
roll-call polling
controller works consecutively through a list of clients, first polling 1, then 2, and so on
Media Access Control (MAC)
controlling when and what computers transmit
media access control (MAC) sublayer
controls physical hardware - how and when bits converted to symbols
WAP (Wireless Access Point)
controls the circuit and determines which client can submit at what time and which circuits have access to mainframe at a given time
MAC provides
data link layer addressing and delimitation of data according to the physical signaling requirements of the medium and type of data link protocol used (implemented in hardware and responsible for data encapsulation)
logical link control (LLC) sublayer
data link layer's connection to the network layer above it
The data link layer provides a means for exchanging
data over a common local medium
Data link layer protocols
described by engineering organizations and communication companies to set a standard
Point-to-Point Protocol
designed to transfer data over this cirucit but provides an address so that it can be used on a multipoint circuit -frame starts with flag, has a 1-byte address, control field is typically not used, protocol field indicates what type of data packet the frame contains -can be up to 1,500 bytes -usually uses CRC-32 but can use 16 too
Addressing fields
destination and source MAC addresses
Error detection
detects errors
Contention-based access
devices can transmit at any time
Bus
devices connect to a common cable
The data link layer provides a means of converting frames into
different network mediums
Multi-access topology
enables a number of devices to communicate by using the same shared media
flow control
ensuring computer sending message is not transmitting too quickly for the receiver
Quality control
flow control services
high-level data link control (HDLC)
formal standard developed by the ISO and often used in WANs --> same as synchronous data control link but can have longer addresses and control fields
Data link layer PDU
frame
The header contains
frame start, addressing fields, type field, quality control
Error control:
handling of network errors caused by problems in transmission
Framing
header, data, and trailer
Star
hosts connect to a common, central device
Logical topologies
how connections appear in the data link layer (determines how the hosts communicate across the medium)
Topology
how the connection between the nodes appears to the data link layer
addressing
identifies source and destination hosts by MAC addresses
type
identifies the layer 3 protocol used by the LLC
Message Delineation
identifying beginning and end of message by using PDU
Octet
in networking, any # that can be represented by 8 bits.
error prevention
includes shielding and moving communication cables away from power cables -can also use amplifiers or repeaters
Crosstalk
is a disturbance caused by the electric or magnetic fields of a signal on one wire to the signal in an adjacent wire.
token passing
one device starts the poll and passes it to the next computer on the multipoint circuit which sends its message and passes the poll to the next -also known as hub polling
Error Detection
is there an error or is there not?
VLAN Virtual LAN
lets you have multiple LANs operating on the same physical equipment. Used to let you separate different types of traffic, like the phones on one LAN, Desktops on another LAN, Printers on a LAN.
The data link layer is divided into two sub-layers
logical link control (LLC) and media access control (MAC)
attenuation
loss of power a signal suffers as it travels from the transmission computer to the receiving computer -caused by gradual decrease in signal over distance -can be prevented by using repeaters or amplifiers
synchronous data control link
mainframe protocol that uses a controlled-access media access protocol (SDLC) -starts with a special bit pattern (01111110) known as a flag -has an address field, control field, information frame, frame check sequence, and a flag
Frame start
marks the beginning of the frame
frame stop
marks the end of a frame
Frame stop
marks the end of the frame
TRIB ( Transmission Rate of Information Bits)
measure of the effective number of bits transmitted in a unit of time
hub polling
one device states the poll and passes it to the next computer on the multipoint circuit
acknowledgement (ACK)
message sent to indicate packet was received without an error
Extended star
multiple stars connected to a single, central networking device
negative acknowledgement (NAK)
negative acknowledgement to indicate message contained an error
Used by an end device to connect to the network
network interface card (NIC)
RJ45
network standard jack
cross-talk
occurs when 1 circuit picks up signal in another -caused by multiplexer guards too small or wires too close -prevented by increase in guard band or moving shielding wires
Single Bit-Error:
one bit of a given data unit (byte, character, packet) is changed from 1 to 0 or from 0 to 1
Hub Polling
one computer starts the poll and passes It to the next on a multipoint circuit, and continues in sequence until the poll reaches the first computer, which then starts the cycle all over again -Token is a unique series of bits
polling
process of sending a signal to a client computer that gives it permission to transmit --> client store all message that need to be transmitted and then controlling device polls the client to see if its been sent
Polling **Exam Question**
process of sending a signal to a client that gives it permission to transmit **Exam Question **
Layer 2 elements
provide an interface, provide an address, uses frames to organize, controls the transmissions of data
Wireless NIC adaptors
provide wireless communication capability to each network host
repeater
receives incoming signal translates it into a digital message and retransmits the message
synchronization
recognition of the start and stop message
media access control
refers to the need to control when computers transmit data --> with point-to-point full duplex configurations, this is necessary with 2 computers -also important with several computers to ensure that no 2 computers attempt to transmit data at the same time and if there is a problem then there must be a way to resolve it
dispersion
refers to the spreading out of a light pulse over time. the more there is, the greater the loss of signal strength
ARP provides two functions
resolving IPv4 addresses to MAC addresses and maintaining a cache of mappings
echo
result of poor connections that cause the signal to reflect back to the transmission equipment -can be prevented by fixing the connections or tuning the equipment
intermodulation noise
special type of cross talk where the signals from 2 circuits combine to form a new signal that falls into a frequency band reserved for another signal -caused by signals from several circuits combined -can be prevented by moving or shielding
control
specifies special flow control services
amplifier
takes the incoming signal, increases its strength and retransmits it on the next section of the circuit
checksum
the "amount" of data sent (calculated with a CRC) which must be equal to the amount of data received (CRC) or it's thrown out as incomplete.
Go-Back-N ARQ
the 1st packet with an error and all of those that follow it
Payload
the actual data being transported, which is everything that is not a header
throughput
the amount of traffic that a network can handle
CSMA/CD (traditional forms of ethernet use this method)
the device listens to the media for the presence of a data signal
CSMA/CD (used by wireless networking technologies)
the device monitors the media for the presence of a data signal
The data link layer is responsible for
the exchange of frames between devices over the media
The data link layer connects the processes of software and protocols to
the hardware of the physical layer
Data
the layer 3 packet inside the frame
digital bandwidth
the measure of data that can flow over a given amount of time
goodput
the measure of useable data transferred over a set amount of time but it doesn't account for network overhead
End finish
the media ends are not well polished, or dirt is present at the termination
Point-to-point topology
the media interconnects just two devices directly
straight-through
the most common type of networking cable. It is used to connect different devices
Physical topology
the physical layout of a network
fiber-optic cables
the signals are patterns of light
wireless
the signals are patterns of microwave transmission
latency
the time that network devices take to look at the data
Link Access Protocol for Modems (LAP-M)
these packets being retransmitted might be the only ones that have an error
copper
these signals are patterns of electrical pulses
controlled access
they control the circuit and determining which clients can transmit at what time
varying the number of twists per pair
to further enhance the cancellation effect of paired circuit wires
Link Layer address can also be called physical address, link address or MAC address t/f
true
t/f Everytime the data packet goes from one device to another, the source and destination MAC address change.
true
Burst Error (More Common):
two or more bits in a data unit are changed at the same time . The length of the ______________ is measured from the first corrupt bit to the last corrupted bit; some in between may not be corrupt.
Type field
type of PDU contained in the frame
checksum
typically 1 byte is added to the end of a message, sum divided by 255, and then using the reminder to check what was transmitting -detects about 95% of error
start bit
typically a binary 0 (in asyn) and is placed at the beginning of a message
stop bit
typically a binary 1 (in asyn) and is placed at the end of a message
RJ11
used for phone equipment
Frame check sequence (FCS)
used to determine if errors occurred in the transmission and reception of the frame
error detection
used to send extra data with each message by the data link layer of the sender on the basis of some mathematical calculations performed on the message
cyclic redundancy check (CRC)
uses a specific calculation using 1 long binary # (P), a whole number (Q), a fixed binary number (G), and a remainder (R) -G is chosen so that the remainder will always be 8, 16, 24, or 32 bits -the remainder is appended to the message as error checking characters -receiving hardware divides the message by G to generate R -has an error detection rate of 99.998%
forward error connection
uses codes containing sufficient redundancy to prevent errors by detecting and correcting them at the receiving end without transmission of the original message
Link Access Protocol-Balanced (LAP-B)
uses same structure as HDLC but is scaled-down version --> ex: it may provide fewer benefits
parity bit
value based on the number of 1s in each byte transmitted --> is set to make the total # of 1s in the whole byte be either even or odd
What are the two solutions for framing? (Shebaro said three but only gave us two)
• Always send a fixed size frame. • Start with a marker so that the receiver will find the start, then followed by the byte count, then the data.
What are the steps for the Hamming Code?
• Put check bits in positions p that are powers of 2, starting with position 1. • Positions → 1 2 3 4 5 6 • 1 is the 1st checkbit, then 2, then 4, etc...
What are the characteristics of parity codes and checksums?
✔ Simple to implement. ✘ Undetected errors are reasonably likely.
