2.

9 Error Detection and Correction

 Network must be able to transfer data from one device to another with
acceptable accuracy.
 For most application, a system must guarantee that the data received are
identical to the data transmitted.
 An electromagnetic signal is subject to interference from heat, magnetism, and
other forms of electricity
 Data can be corrupted during transmission and some applications require that
errors be detected and corrected.
 Some application can tolerate a small level of error – random errors in audio or
video
 Some need very high level of accuracy - text

Chapter 2 : Data Communications 1

2.9 Error Detection and Correction
 Types of Errors

 Single-bit Error
 Single-bit error: 0  1 or 1  0
 Only one bit of a given data unit is changed
 The least likely type of error in serial transmission. Why ?
 Single-bit error can happen in parallel transmission

Chapter 2 : Data Communications 2

2.9 Error Detection and Correction
 Single-bit Error

 Burst Error
 Two or more bits in the data unit have changed
 Burst error does not necessarily mean that the errors occur in consecutive bits
 Most likely to happen in a serial transmission
 Number of bits affected depends on the data rate and duration of noise

Chapter 2 : Data Communications 3

2.9 Error Detection and Correction
 Burst Error
 Ex : If the data is sent at 1 kbps, a noise of 1/100 s can affect 10 bits.
 Ex : If the data is sent at 1 Mbps, a noise of 1/100 s can affect 10000 bits.

Chapter 2 : Data Communications 4

2.9 Error Detection and Correction
 Redundancy
 Error detection uses the concept of redundancy, which means adding extra
(redundant) bits for detecting errors at the destination

Chapter 2 : Data Communications 5

2.9 Error Detection and Correction
 Detection vs. Correction
 Detection: error ? yes or no
 Correction: Need to know the exact number of bits that are corrupted, and their
location in the message
 Ex : one error in 8 bits sequence – 8 possible location.
 Ex : 10 errors in 1000 bits sequence ?

 Forward Error Correction vs. Retransmission

 2 main methods of error correction :
 Forward Error Correction – the receiver tries to guess the message by using
redundant bits (in the case the number of errors is small).
 Retransmission – the receiver detects the occurrence of an error and ask the sender
to resend the message. This process is repeated until a message arrives that the
receiver believes is error-free.

Chapter 2 : Data Communications 6

2.9 Error Detection and Correction
 Hamming Distance
 The Hamming distance between two words is the number of differences between
corresponding bits.
 Easily found by applying the XOR operation on the two words and count the number
of 1s in the result.
 Ex : Hamming distance d (000,011)
 Ex : Hamming distance d (10101, 11110)

 Minimum Hamming Distance

 The minimum Hamming distance is the smallest Hamming distance between all
possible pairs in a set of words
 Ex : Find the minimum Hamming distance of the coding scheme in below table.

Chapter 2 : Data Communications 7

2.9 Error Detection and Correction
 Minimum Hamming Distance
 Ex : Find the minimum Hamming distance of the coding scheme in below table.

 3 parameters to define the coding scheme

 Codeword size n
 Dataword size k
 The minimum Hamming distance dmin

Chapter 2 : Data Communications 8

2.9 Error Detection and Correction
 Relation between Hamming Distance and Error
 When a codeword is corrupted during transmission, the Hamming distance between
the sent and received codewords is the number of bits affected by the error.
 In other words, the Hamming distance between the received codeword and the sent
codeword is the number of bits that are corrupted during transmission.
 Ex : if the codeword 00000 is sent and 01101 is received, 3 bits are in error and the
Hamming distance between the two is d (00000, 01101) = 3.

 Minimum Distance for Error Detection

 If s errors occur during transmission, the Hamming distance between the sent
codeword and the received codeword is s.
 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.
 Examples

Chapter 2 : Data Communications 9

2.9.1 Error Detection & Correction Common Techniques
 Simple Parity-Check Code
 A k-bit dataword is changed to an n-bit codeword where n = k + 1.
 The extra bit (called the parity bit), is selected to make the total number of 1s in the
codeword even.
 The minimum Hamming distance dmin = 2, which means that the code is a single-bit
error-detecting code.
 Example of parity-check code with k = 4, n = 5; C (5,4) :

Chapter 2 : Data Communications 10

2.9.1 Error Detection and Correction Common Techniques
 Encoder and Decoder for Parity-Check Code

Chapter 2 : Data Communications 11

2.9.1 Error Detection & Correction Common Techniques
 Encoder and Decoder for Parity-Check Code
 The encoder uses a generator that takes a copy of a 4-bit dataword (a0, a1, a2 and
a3) and generates a parity bit r0.
 The dataword bits and the parity bit create the 5-bit codeword.
 The parity bit that is added make the number of 1s in the codeword even.
 This is done by adding the 4 bits of the dataword (modulo-2) where the result is the
parity bit.
r0 = a3 + a2 + a1 + a0

 The sender sends the codeword which may be corrupted during transmission and the
receiver receives a 5-bit word.
 The checker at the receiver does the same thing as the generator at the sender, but
the addition is done over all 5 bits.

s0 = b3 + b2 + b1 + b0 + q0

Chapter 2 : Data Communications 12

2.9.1 Error Detection & Correction Common Techniques
 Encoder and Decoder for Parity-Check Code
 The result which is called the syndrome, is just 1 bit.
 The syndrome is passed to the decision logic analyzer.
 If the syndrome = 0 = no error in the received codeword; the data portion of the
received codeword is accepted as the dataword.
 If the syndrome = 1 = error in the received codeword; the data portion of the
received codeword is discarded and no dataword is created.

Chapter 2 : Data Communications 13

2.9.1 Error Detection & Correction Common Techniques
 Two-dimensional parity-check code

Chapter 2 : Data Communications 14

2.9.1 Error Detection & Correction Common Techniques
 Two-dimensional parity-check code

Chapter 2 : Data Communications 15

2.9.1 Error Detection & Correction Common Techniques
 Error Correcting Code : Hamming Codes C (7,4)

Chapter 2 : Data Communications 16

2.9.1 Error Detection & Correction Common Techniques
 Encoder and decoder for a Hamming Code

Chapter 2 : Data Communications 17

2.9.1 Error Detection & Correction Common Techniques
 Logical Decision made by the correction logical analyzer of the decoder :

 Examples

Chapter 2 : Data Communications 18

2.9.1 Error Detection & Correctioon Common Techniques
 The use of Hamming Code in burst error correction
 A Hamming code can only correct a single error and detect a double error.
 But there is a way to make it detect a burst error :

Chapter 2 : Data Communications 19

2.9.1 Error Detection & Correction Common Techniques
 The use of Hamming Code in burst error correction

 When a burst of error of size 4 corrupts the frame, only 1 bit from each codeword is
corrupted – the corrupted bit can be easily be corrected at the receiver.

Chapter 2 : Data Communications 20

2.9.1 Error Detection & Correction Common Techniques
 Cyclic Codes
 Cyclic codes are special linear block code – if a codeword is cyclically shifted
(rotated), the result is another codeword
 Ex : 1011000 is a codeword and if cyclically left-shifted, the result 0110001 is also a
codeword.

Chapter 2 : Data Communications 21

2.9.1 Error Detection & Correction Common Techniques
 Cyclic Redundancy Check
 The use of cyclic codes to detect and correct error
 CRC encoder and decoder

Chapter 2 : Data Communications 22

2.9.1 Error Detection & Correction Common Techniques
 CRC encoder and decoder
 In the encoder, the dataword has k
bits (4 bits)
 The size of dataword is augmented
(increased) by adding n – k (3 bits)
0s to the right-hand side of the word.
 The n-bits is then fed to the
generator.
 The generator uses a pre-defined
divisor to divide the dataword
(modulo-2 division).
 The quotient of the division is
discarded, the remainder (r2r1r0) is
appended to the dataword to create
the codeword.

Chapter 2 : Data Communications 23

2.9.1 Error Detection & Correction Common Techniques
 CRC encoder and decoder
 The decoder receives the possibly
corrupted codeword.
 A copy of all n bits is fed to the
checker to do the same step as the
generator.
 The remainder produced by the
checker is a syndrome of n – k (3
bits) which is fed to the decision
logic analyzer.
 If the syndrome bits are all 0s, the 4
leftmost bits of the codeword are
accepted as a dataword.
 Otherwise, the 4 bits are discarded
(error).

Chapter 2 : Data Communications 24

2.9.1 Error Detection & Correction Common Techniques
 Example : Division in CRC Encoder

Chapter 2 : Data Communications 25

2.9.1 Error Detection & Correction Common Techniques
 Example : Division in CRC Decoder

Chapter 2 : Data Communications 26

2.9.1 Error Detection & Correction Common Techniques
 Checksum
 The idea as in ex 10.18.
 It can made easier as in ex 10.19

Chapter 2 : Data Communications 27

2.9.1 Error Detection & Correction Common Techniques
 Actual implementation of checksum
 Ex : The sender initializes the checksum to 0 and adds all data items and the
checksum. However, 36 cannot be expressed in 4 bits. The extra two bits are
wrapped and added with the sum to create the wrapped sum value 6. The sum is
then complemented, resulting in the checksum value 9 (15 − 6 = 9).

Chapter 2 : Data Communications 28

2.9.1 Error Detection & Correction Common Techniques
 Internet checksum
 Sender site:
 1. The message is divided into 16-bit words.
 2. The value of the checksum word is set to 0.
 3. All words including the checksum are added using one’s complement
 addition.
 4. The sum is complemented and becomes the checksum.
 5. The checksum is sent with the data.
 Receiver site:
 1. The message (including checksum) is divided into 16-bit words.
 2. All words are added using one’s complement addition.
 3. The sum is complemented and becomes the new checksum.
 4. If the value of checksum is 0, the message is accepted; otherwise, it is
rejected.

Chapter 2 : Data Communications 29

2.9.1 Error Detection & Correction Common Techniques
 Internet checksum : Examples

Chapter 2 : Data Communications 30