ISDS 409

acknoledgment

ACK

data link protocol

Controls when computers transmit (media access control) Detects and correct errors ( error control) Identifies the start and end of a message by using a PDU (message delineation)

Point to point protocal (PPP)

Developed in early 90s and is often used in WANs Commonly used on dial-up lines from home PCs Designed mainly for point-to-point phone line (can be used for multipoint lines as well) Flag 1B Address 1B Control 1B Protocol 2B Data Variable length Frame check sequence 2B or 4 B Flag 1B

which is better

in general contention approach works better than access control, for small networks that have low usage. access control works better for large networks with high usage.

LAP-B (Link Access Protocol - Balanced)

uses same structure as HDLC but is scaled down.

Ideal frame size

2000 to 10000 bytes

Transport Layer Functions

Linking to Application Layer Segmenting Session Management Connection-oriented messaging Connectionless messaging Quality of Service (QoS) routing

negative acknowledgment

NAK

forward effer correction

Uses codes containing sufficient redundancy to prevent errors by detecting and correcting them at the receiving end without re transmission of the original message. Commonly used in satellite transmission, due to high satellite cost, cost of forward error correction is insignificant.

synchronous transmission

all letters or data in a group are transmitted at one time as a block of data called a frame. Used in point to point of multi point circuits. For multi point circuits each packet must contain a destination and source address.

TCP/IP

is the most commonly used set of transport and network later protocols.

Polling

is the process of sending a signal to a client computer that gives it permission to transmit. With polling clients store all messages that need to be transmitted. Periodically the controlling device polls the client to see if it has any data to transmit if it does, then it transmits.

Transport layer

links the application software in the application layer with the network layer andd is responsible for the end to end delivery of the message. Responsible for segmenting and reassembly Breaking the message into several smaller pieces at the source Reconstructing the original message into a single whole at the destination The PDU of the transport layer is a segment.

Automatic Repeat reQuest (ARQ)

re transmission Stop and wait continuous.

Network layer

the network layer takes the message and routes it through the network by selecting the best path from computer to computer. Responsible for addressing and routing of messages Selects the best path from computer to computer until the message reaches destination Adds IP packet

synchronization

the recognition of the start and stop of each message.

Check sum

1 byte added to the end of the message. 95% error detection Calculated by adding decimal values of each character in the message Dividing the total by 255 and Saving the remainder (typically 1 byte) and using it as the checksum

IPv6 Addresses

128 bit (16 Byte) addresses Source address field has 128 bits Destination address field has 128 bits Hexadecimal (base 16): Similar to Ethernet addresses Breaks the address into 8 blocks of 2-byte numbers and writes the hexadecimal equivalent for each 2-byte number Example: 2001:0890:0600:00d1:0000:0000:abcd:f010 "Compressed notation" eliminates leading zeros within each block and blocks that are all zeros Example: 2001:890:600:d1: :abcd:f010

cross over point

20 computers, lower for busier computers higher for less busy computers.

IPv4 Addresses

32 bit (4 byte) addresses Source address field has 32 bits Destination address field has 32 bits Dotted decimal notation: Used to make IP addresses easier to understand for humans Breaks the address into 4 Bytes and writes the decimal equivalent for each byte Example: 128.192.56.1

transmission efficiency

= information bits/ (information bits + overhead bits)

High-Level Data Link Control (HDLC)

Formal standard developed by ISO often used in WAN. Uses controlled access MAC Same as SDLC, except Longer address and control fields Larger sliding window size

data link layer PDU added to the network layer PDU

Frame

IPv4 and IPv6 Address Space

IPv4 uses 32 bit addresses—theoretical maximum of 4.2 billion addresses 500 million of these addresses are reserved IPv4 address space has been used up quickly Will have run out of IPv4 addresses in 2011 IPv6 uses 128 bit addresses—theoretical maximum of 3.4 x 1038 addresses, a very large number IPv6: Mainly developed to increase IP address space due to the huge growth in Internet usage (128 bit addresses) Little chance this address space will ever be used up

Synchronous Data Link Control (SDLC)

IS a mainframe protocal developed by IBM. Uses a controlled access media access protocol. Flag 8 bits Address - identifies destination 8 bits Control- identifies the kind of frame that is transmitted, information(messages) or supervisory(ACK and NAK). 8 bits Message - variable length contains message. Frame Check Sequence - is a 32 bit crc code Flag 8 bits

Cisco HDLC

Includes a network protocal field. cHDLC and HDLC have gradually replaced SDLC

Transmission Control Protocol / Internet Protocol (TCP/IP)

Most common, used by all Internet equipment Developed in 1974 by V. Cerf and B. Kahn As part of ARPANET (U.S. Department of Defense) Most common protocol suite Used by the Internet Used by almost all backbone and wide area networks Most commonly used protocol on LANs Reasonably efficient and error free transmission Performs error checking Transmits large files with end-to-end delivery assurance Compatible with a variety of data link layer protocols

Ethernet formal standard

Most widely used LAN protocol, developed jointly by Digital, Intel, and Xerox Uses contention-based MAC Two standard versions of Ethernet IEEE 802.3ac 33-34 bits of overhead If VLAN isnt used ithe field is omitted. 29-30 bits. Ethernet II 26 bits of overhead

Ethernet II

Preamble (7B): Start of Frame (1B): Addresses (6B each) Type (2B): specify an ACK frame or the type of network layer packet the frame contains (e.g., IP packet) Frame Check Sequence (4B): CRC-32

Ethernet IEEE 802.3ac Frame

Preamble (7B): repeating 1s and 0s Start of Frame (1B): marks beginning of frame destination Address (6B) Source address Address (6B) VLAN Tag (4B): only exists when VLAN is used. If VLAN is used, first 2B are set to 24,832 (8100H) Length (2B): shows the number of bytes in the data field (43B-1497B) DSAP (1B) SSAP (1B): Pass control info between sender and receiver (e.g., shows what network layer protocol the frame contains) Control (1-2B): holds frame sequence numbers, ACKs and NAKs, and other control info Data (46-1500B) Frame Check Sequence (4B): CRC-32

User Datagram Protocol (UDP)

Protocol used when sender needs to send a single small packet to the receiver (e.g., DNS request) - Faster Uses a small header - only 8 bytes containing 4 fields: Source port Destination port Length CRC-16 UDP does not check for lost messages (unlike TCP) UDP is used for connectionless messaging in TCP/IP suite (no ACKs, no flow control) Commonly used for control messages that are usually small (e.g., DNS) Can also be used for applications where a packet can be lost, such as information-rich video PDU are called diagrams. Engineer decides when to use TCP or UDP is going to be used.

Error prevention

Shielding wire.(more expensive and harder to install) Moving cables away from source of noise. (power sources) FDM to TDM or changing frequencies or guard bands in FDM can help. (cross talk and inter modulation) Tune the transmission and redo the connections. (echos and white noise)

Linking to Application Layer

TCP may serve several Application Layer protocols at the same time Problem: Which application layer program to send a message to? Solution: Port numbers located in TCP header fields; 2-byte each (source, destination) Standard port numbers Standardized port addresses Nonstandard port numbers Possible, but requires configuration of TCP Can be used to enhance security from commonly known ports source port address and destination port address.

Internet Protocol (IP)

Two versions in current in use IPv4: Header has 192 bits (24 Bytes); it uses 32 bit addresses. If Option" field is not used, then header has 160 bits (20 Bytes). IPv6: Header has 320 bits (40 Bytes); it uses 128 bit addresses. Might run out of IP addresses in 2011 Iv6 has bigger address capacity.

TCP

Typical segment is 192 buts (24 bytes) Links the application layer to the network layer Performs segmenting and reassembly Breaking up a large message into smaller segments Numbering the segments, and Reassembling them at the destination Ensures reliable delivery of packets If no Options field, TCP header is 160b (20B) long optional and rarely used.

Transmission Rate of Information Bits (TRIB)

Used as a measurement of throughput = Number of info bits accepted / total time required to get the bits (Number of info bits) *(Prob. of successful transmission) / (Time it takes to transmit these bits) + (Propagation delay

Echoes

a result of poor connection that caused the signal to reflect back to the transmitting equipment. can also be heard in fiber cables if not properly aligned.

cyclic redundancy check

adds 8,16,24, or 32 bits to the message. Treated as one long binary number. About 100% error detection. P / G = Q + R / G Detects 100% of errors (if number of errors <= size of R) Otherwise: CRC-16 (99.998%) and CRC-32 (99.99999998%)

line noise and distortion

can cause data communication errors. Noise can manifest its self as extra bits, missing bits or flipped bits.

white noise or gaussian noise

caused by thermal agitation of electrons and therefore inescapable. even if cable is perfectly insulated there would still be some white noise. solutions increase strength of electrical signal to over power the white noise. Static sound

repeaters

commonly used in digital circuits. A repeater receives the incoming signal, translates it into a digital message and re transmits the message. Noise and distortion is not amplified. Cleaner lower error rate transition.

Contention

computers wait until the circuit is free and then transmit the data. Commonly used in Ethernet LANs

network errors

controlled by the network hardware and software.

human error

controlled through the application program

Controlled access

controls the circuit and determines which client can transmit at what time. Access request client computers that want to transmit send a request to transmit to the device that is controlling the circuit (wireless AP). The controlling device grants permission for one computer to transmit at a time.

media sublayer control sublayer (MAC)

controls the physical hardware. At the sending computer, it controls how and when the physical layer converts bits into physical symbols that are sent down th circuit. At the sending end, the MAC sub layer 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. At the receiving end, the MAC receives a stream of bits from the physical layer and translates it into a coherent PDU, ensures that no errors have occurred during transmission and passes the data link PDU to the LLC. both sender and receiver have to agree on the rules or protocols

type of network error

corrected data (data that has been changed) lost data Networks should be designed to prevent, detect and correct both corrupted and lost data.

Flow control

ensuring that the computer sending the messages is not transmitting too quickly for the receiver. By using ACK and NAK the receiver can control the rate at which it receives information. Stop and wait, receiver doesn't send ACK until it is ready to receive more packets. Continuous, sender and receiver usually agree on the size of the sliding window.

inter modulation noise

is 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.

error detection

is extra data send with each message. error detection increases throughput decreases.

Logical link control sublayer

is the data links layer's connection to the network layer above it. At the sending computer the LLC is responsible for communicating with the network layer and for taking the network layer PDU and surrounding it with a data link layer PDU usually an ethernet frame. At the receiving end the LLC is removes the LLC PDU and passes the message it contains to the network layer.

Attenuation

is the loss of power a signal suffers as it travels from the transmitting computer to the receiving computer. some power is lost or absorbed before it reaches the receiver.

impulse noise (spikes)

is the primary source of errors in data communication. doesnt really affect voice communication but can obliterate a group of data. causing a burst error. Caused by voltage changes, lighting, florescent lights, and poor connections in a circuit. click or cracking noise and can last as long as 1/100 of a second

burst error

more than one bit is changed by the error causing condition. (normally) errors are not uniformly distributed in time. 1 in ever 50,000 most likely to be 100 in ever 50,000,000

Cross talk

occurs when one signal pick ups signals in another. normally not bother some.

Parity Checking

one additional bit is added for every 1 byte. Either even or odd number of 1s. 50%b error detection.

hub polling ( token passing)

one device starts the poll and passes it to the next computer on the multi point circuit which sends its message and passes the poll to the next . That computer then passes the poll to the next until it completes the cycle.

Go back n ARQ

re transmission of the first packet containing an error and all the one that followed

media acces control

refers to control when computers transmit. With point to point full duplex configs media access control is unnecessary because there are only 2 computers in the circuit, and full duplex permits either computer to transmit at the same time. becomes important when there are several computers on the same communication circuit.

stop and wait ARQ

sender stops and waits for a response from the reciever after each data packet. Half duplex technique.

The data link layer is responsible for

sending and receiving messages to and from other computers.

asynchronous transmission

start stop transmission. Used in point to point full duplex transmissions. Start(0)and stop(1) bit added to each individual character. Start and stop transmission.

how to avoid attenuation

telephone circuits have repeaters and amplifiers spaced through out their length.

roll call polling

the controller works consecutively one after the other in order but can prioritize by repeating a client in a sequence.

Continuous ARQ

the sender does not wait for acknowledgment from receiver after every message. It continues to transmit data. Full duplex connection. Sometimes called sliding window.

throughput

the total number of information bits received per second after taking into account the overhead bits ad the the need to re transmit frames containing errors.

Data link layer has two main tasks

theresfore it is often divided into 2 sub layers. Logical link control (LLC) Media access control (MAC)

Link access control protocol for modems (LAP-M)

transmission of only the packets containing errors. More efficient.

amplifiers

typically used in analog circuits. Takes income signal increases its strength and re transmits it. Noise is also amplified.

Error rates

x bit in error for ever xxxxxx bits transmitted.