CENG415 – Communication Networks

1. CENG415 – Communication Networks Lectures 18 Data Link layer – Physical addressing – ARP – Error detection

2. Introduction • data-link layer has responsibility of transferring datagram from one node to adjacent node over a link • Nodes: Hosts and routers • Links: communication channels that connect adjacent nodes along communication path • Layer 2 packet is a frame. The frame encapsulate the datagram. • Links are different (Ethernet, Frame Relay, 802,…) • Different link protocols are used • Each link protocol provides different services

3. Data Link Layer - Services • The DLL receives service from physical layer and provides service to the network layer • encapsulate datagram into frame, adding header, trailer • Responsible for carrying data from one hop to the next hop

4. Hop to hop delivery

5. Addressing • Application layer uses specific addresses • Transport layer uses Port addresses • Network layer uses logical addresses (IP) • Link layer uses physical addresses

6. Network Interface Card • NIC: Network Interface Card • A network interface card is used to connect a device to a network • The card provides an interface to the media In an Ethernet Network: • Each NIC is assigned an Ethernet source address by the manufacturer (stored in a PROM on the card). • MAC address allocation administered by IEEE • manufacturer buys portion of MAC address space (to assure uniqueness) NOTE: MAC Address is called physical address

7. Physical address • Most local-area networks use a 48-bit (6-byte) physical address written as 12 hexadecimal digits; every byte (2 hexadecimal digits) is separated by a colon • Example: 07:01:02:01:2C:4B On windows system, you can perform the command GETMAC to get the MAC addresses (physical addresses of your computer)

8. Link between Hops • Frame travels between Hops • Frame holds the Datagram • Datagram holds the segment • Segment holds the message • The IP address (logical address) is used to simulate all the Internet as homogeneous network • But the Internet is NOT • It is a collection of heterogeneous networks (separated by routers) • Within one network (LAN) the hardware is similar • The hardware understands MAC addresses only

9. ARP: Address Resolution Protocol Problem • A wants to send datagram to B • A knows the IP of B but not its MAC address Solution • A broadcasts ARP query packet containing B’s IP Address. • Broadcasting MAC address in a LAN is FF:FF:FF:FF:FF:FF • All machines on the LAN receive ARP query • Only B reply directly to A with B’s MAC address • B can send the reply directly to A (unicast) because the request of A contains A’s MAC address • A caches (IP ,MAC) pair in its ARP table until information becomes old (times out 20 minutes). Information that times out are lost • ARP is “plug-and-play”: nodes create their ARP tables without intervention from net administrator

10. Windows ARP command Open a command prompt window and type arp /? • ARP command is used to display and modifie the IP-to-Physical address translation tables used by address resolution protocol (ARP) IP broadcasting address for this network MAC broadcasting address in a LAN

11. ARP in depth Consider the following LAN At startup, all ARP tables are empty Scenario 1 A broadcast an ARP request with IP of PC4 The broadcast will reach: • Router 1: You are requesting a MAC of an IP on the same network this is not my business • C2: This is not my IP so I will not reply. But I will add A’s IP and A’s MAC to my ARP table • C3: will do the same as C2 • C4: its me I will send an ARP reply directly to A with my MAC address. I will also add A’s IP and A’s MAC to my ARP table

12. ARP in depth Consider the following LAN At startup, all ARP tables are empty Scenario 2 A is looking for an IP that is not on the same network • A knows that the IP is not on its network • A knows that only the default gateway can help • A broadcast an ARP request with IP of Router1 • The broadcast will reach: • Router 1: You this is my MAC I will send it to you directly. And I will add your IP and MAC address to my ARP table • C2: This is not my IP so I will not reply. But I will add A’s IP and A’s MAC to my ARP table • C3 and C4 : will do the same as C2

13. LAB: ARP • We will use the packet tracer network created in lecture 16 • Open PC0 Command prompt and type arp –a • If you have entries in the ARP table clear them using arp -d • Switch to simulation mode • Type ping 192.168.1.4 (PC4) • Open the envelop (ARP) • Notice the following entries

14. LAB: ARP • Click on Capture and Forward once. The Packet reaches the Switch. • Click on Capture and Forward again. The packet is sent to all other devices. • They all discard it except PC4 • Click on Capture and Forward again • Check the reply (unicast)

15. ARP message format • Hardware type (HRD) • Protocol type: This field specifies the type of layer three addresses used in the message. For IPv4, this value is 2048 (0800 hex) EtherType code for the Internet Protocol • Hardware Address Length: Specifies how long hardware addresses are. The value is 6 for Ethernet or any IEEE 802 MAC addresses • Protocol Address Length: the complement of the precedingfield; specifies how long protocol (layer three) addresses are in this message. For IP(v4) addresses this value is 4

16. ARP message format • Opcode: this field specifies the nature of the ARP message

17. frame frame Adapter communicating datagram • sending side: • Find the corresponding MAC of the receiving node • encapsulates datagram in a frame • adds error checking bits, rdt, flow control, etc receiving node link layer protocol sending node adapter adapter • receiving side • looks for errors, rdt, flow control, etc • extracts datagram, passes to receiving node • adapter is semi-autonomous

18. Error Detection • The objective is to detect transmission errors • With a possibility to correct errors • Error detection not 100% reliable! • protocol may miss some errors, but rarely • Frames are transmitted as follow • EDC= Error Detection and Correction bits (redundancy) Appended to the message • D = Data protected by error checking, may include header fields NOTE: larger EDC field yields better detection and correction

19. Error Detection and Correction

20. Parity Check – Single bit parity • Single bit parity: detects single bit error • Parity bit can be: • Even: the number of ones, including parity bit should be even Example: 01110110001  even parity bit is 0 11110011010  even parity bit is 1 • Odd: the number of ones, including parity bit should be odd Example: 01110110001  odd parity bit is 1 11110011010  odd parity bit is 0 Problems with parity bit. If two bits flip, error is not detected Even parity Original message 01110110001 received message 010101100010  error Original message 01110110001 received message 010101000010  No error

21. Single bit parity design

22. Hamming Distance We define: • Hamming distance: d(x, y) between two words x and y (of the same size) is the number of differences between the corresponding bits (XOR) Example: d(000, 011) = 2 since 000 ⊕ 011 = 011 (two 1s) d(10101, 11110) = 3 since 10101 ⊕ 11110 = 01011 (three 1s) • Minimum Hamming distance: is the smallest Hamming distance between all possible pairs in a set of words Example: Find the minimum Hamming distance of the words in the following table Solution: dmin = 3

23. Hamming Distance • To guarantee the detection of up to s errors in all cases, the minimum Hamming distance in a block code must be dmin= s + 1. • To guarantee correction of up to t errors in all cases, the minimum Hamming distance in a block code must be dmin = 2t + 1. Example: A code has a Hamming distance dmin = 4. What is the error detection and correction capability the code? Solution: This code guarantees the detection of up to three errors (4 = s + 1  s = 3), but it can correct up to one error (4 = 2t + 1  t = 3/1 = 1).

24. Block Coding In block coding, we divide our message into blocks, each of k bits, called datawords. We add r redundant bits to each block to make the length n = k + r The resulting n-bit blocks are called codewords. • With k bits, we can create a combination of 2Kdatawords • With n bits, we can create a combination of 2ncodewords Since n > k, the number of possible codewords is larger than the number of possible datawords. The block coding process is one-to-one; the same dataword is always encoded as the same codeword. This means that we have 2n− 2K codewords that are not used. We call these codewords invalid or illegal.

25. Block Coding Error detection Error correction

26. Linear Block Coding • In a linear block code, the exclusive OR (XOR) of any two valid codewords creates another valid codeword • The minimum Hamming distance is the number of 1s in the nonzero valid codeword with the smallest number of 1s Example: A simple parity-check code is a single-bit error-detecting code in which n = k + 1 with dmin =2 The Syndrome is the addition (⊕) of thecodeword bits Syndrome = 0meaning that the codeword is OK

27. Linear Block Coding

28. Cyclic Code Cyclic codes are linear codes with one extra property, if a codeword is cyclically shifted (rotated), the result is another codeword. In this lecture, we will discuss a category of cyclic codes: the cyclic redundancy check (CRC) used in networks such as LANs and WANs Example: CRC Code with C(7, 4) n=7 k = 4

29. Cyclic Code

30. Cyclic Code - Encoder In the encoder, the dataword has k bits (4); the codeword has n bits (7). The size of the dataword is augmented by adding n − k (3) 0s to the right-hand side of the word. The n-bit result is fed into the generator. The generator uses a divisor of size n − k + 1 (4), predefined and agreed upon. The generator divides the augmented dataword by the divisor (modulo-2 division). The quotient of the division is discarded; the remainder (r2r1r0) is appended to the dataword to create the codeword.

31. Cyclic Code - Decoder The decoder receives the possibly corrupted codeword. A copy of all n bits is fed to the checker which is a replica of the generator. The remainder produced by the checker is a syndrome of n − k (3) bits, which is fed to the decision logic analyzer. The analyzer has a simple function. If the syndrome bits are all 0s, the 4 leftmost bits of the codeword are accepted as the dataword otherwise, the 4 bits are discarded

32. CRC Encoder – Example

33. CRC Decoder – Example

34. Using polynomial A binary number can be converted to a polynomial as follow: If there is a 1 in the position i of the binary number add xi-1 to the polynomial Example: 110101  x0 + x2 + x4 + x5 = x5 + x4 + x2+ 1 1001000  x6+ x3 Dataword in the previous example 1011  x3+ x1+ 1 Divisor in the previous example = 1001110

35. Standard polynomials In a LAN • The sender computes CRC of the message and appends the computed CRC at the end of message • The receiver computes CRC on the message and the received CRC, if result is 0 message is OK This table shows some standard polynomials used in Networking

36. Internet Checksum Sender: • treat segment contents as sequence of 16-bit integers • checksum: addition (1’s complement sum) of segment contents • sender puts checksum value into UDP checksum field Receiver: • compute checksum of received segment • check if computed checksum equals checksum field value: • NO - error detected • YES - no error detected.