Errors, Error Detection, and Error Control

1
Data Communications and Computer Networks A
Business Users Approach Third Edition

• Chapter 6
• Errors, Error Detection, and Error Control

2
Topics

• Different types of noise found in computer
  networks
• Different error-prevention techniques for
  different types of noise
• Efficiency of error-detection techniques
• Parity calculations
• Cyclic redundancy checksum (CRC)
• Types of errors detected CRC
• Error control and its applications
• Stop-and-wait ARQ, Go-back-N ARQ, and
  Selective-reject ARQ

3
Introduction

• Noise is always present on circuits
• Too much noise can corrupt signals and the
  associated data
• If an error occurs a system may either request
  retransmission of the packet in error or drop the
  corrupted packet

4
White Noise

• Also known as thermal or Gaussian noise
• Relatively constant
• Can be reduced by conditioning circuits
• White noise can completely disrupt a signal

5
White Noise Illustrated

6
Impulse Noise

• One of the most disruptive forms of noise
• Random spikes of power
• Can destroy one or more bits of information
• Difficult to remove from an analog signal
• May be hard to distinguish from original signal
• Impulse noise can damage more bits if the bits
  are closer together (transmitted at a faster rate)

7
Impulse Noise Illustrated

8
Impulse Noise and Transmission Speed

9
Crosstalk

• Unwanted interference of signals on two different
  signal paths
• Example - hearing another conversation while
  talking on the telephone
• Relatively constant
• Can be reduced by adding shielding and/or more
  twists per inch to conductor

10
Crosstalk Illustrated

11
Echo

• Signal rebound feedback
• Most often occurs on coaxial cable
• Could interfere with original signal
• Relatively constant
• Can be significantly reduced by properly
  terminating and grounding conductors

12
Echo Illustrated

13
Jitter

• The result of small timing irregularities during
  transmission of digital signals
• Occurs when a digital signal is repeated over and
  over
• May force the system to slow down its
  transmission
• Can be reduced by properly conditioning circuits

14
Jitter Illustrated

15
Delay Distortion and Attenuation

• Delay Distortion
• occurs because the velocity of propagation of a
  signal through a medium varies with the frequency
  of the signal
• Can be reduced
• Attenuation
• Loss of a signal strength as the signal travels
  through a medium

16
Error Prevention

• Error Prevention Techniques
• Proper shielding of cables to reduce interference
• Telephone line conditioning or equalization
• Replacing older media and equipment with digital
  components
• Proper use of digital repeaters and analog
  amplifiers
• Observe the stated capacities of the media

17
A Summary of Error Prevention Techniques

18
Error Detection

• Despite best prevention techniques, errors still
  occur
• Error detection code are added to the data/signal
  to detect errors
• Two basic techniques for detecting errors
• Parity checking
• Cyclic redundancy checksum

19
Parity Checks

• Simple parity
• even parity adds a bit to make the number of 1s
  an even number
• odd parity adds a bit to make the number of 1s an
  odd number
• 1001010 using even parity is transmitted as
  10010101

20
Parity Checking Drawback

• If two bits are flipped during transmission the
  errors will go undetected
• Simple parity only works if an odd numbers of
  bits are transposed during transmission

21
Longitudinal Parity

• Vertical and horizontal parity bits added to a
  block of characters
• Vertical bits are simple parity bits
• Horizontal bits are column checks
• Extra bits add considerable redundancy to a
  character block

22
Longitudinal Parity Illustrated

23
Longitudinal Parity May Miss Errors

24
Parity Issues

• Both simple and longitudinal parity do not catch
  all errors
• Simple parity only catches odd numbers of bit
  transitions
• Longitudinal parity is better but adds a lot of
  extra check bits to the block of data

25
Cyclic Redundancy Checksum

• CRC error detection associates the data packet
  with as a large polynomial
• The packet transmitter divides the polynomial by
  a given generating polynomial and saves the
  remainder
• This remainder is attached to the end of message

26
CRC Recalculation

• The message is transmitted to the receiver
• The receiver calculates the remainder for the
  received message
• If the two remainders match, then message is
  considered error-free (See table next slide)

27
Error Detection Rates for CRC

28
Error Control

• Once an error is detected the receiver can
• Do nothing
• Return an error message to the transmitter
• Fix the error if redundant data are provided

29
Error Control Strategies

• Do nothing
• Packets are dropped
• May make little difference in quality of
  transmission (streaming video)
• Some digital systems rely on low error rates
  (frame relay)

30
Retransmission Error Control

• Return a message for retransmission of data in
  error
• Stop-and-wait ARQ
• Go-back-N ARQ
• Selective-reject ARQ

31
Stop-and-wait ARQ

• Simplest error control protocol
• A transmitter sends a frame then stops and waits
  for an acknowledgment
• If a positive acknowledgment (ACK) is received,
  the next frame is sent
• If a negative acknowledgment (NAK) is received,
  the same frame is transmitted again

32
Stop-and-wait ARQ Illustrated

33
Go-back-N ARQ

• Go-back-N ARQ and selective reject are more
  efficient because multiple frames are sent using
  a sliding window protocol
• A sliding window protocol allows a transmitter to
  send a block of frames before an acknowledgment
  is returned
• Acknowledgments are accompanied by a number
  indicating how many frames can be sent in the
  next exchange

34
Sliding Window Protocol

35
Error Handling in Go-back-N ARQ

• If a frame is received with errors, the receiver
  asks the transmitter to retransmit all frames
  from a certain point on
• Transmitting station may sent a request if
  receiver is too slow in responding

36
Selective-reject ARQ

• Most efficient error control protocol
• If a frame is received in error, the receiver
  asks transmitter to resend ONLY the frame that
  was in error
• Subsequent frames following the Nth frame are not
  retransmitted
• Transmitting station may sent a request if
  receiver is too slow in responding

37
Transfer Without Errors

38
Go-back N ARQ with Packet Loss

39
Selective-reject ARQ with Packet Loss

40
Correct the Error

• The receiver corrects the error with no help from
  the transmitter
• Requires a large amount of redundant information
  accompanying original data
• This redundant information allows the receiver to
  determine the error and make corrections
• This type of error control is often called
  forward error correction (FEC)

41
Error Detection in Action

• Asynchronous transfer mode (ATM) detects and
  controls certain types of errors
• ATM uses CRC to check for errors in the frame
  header
• ATM CRC is powerful enough to perform simple
  error correction on the header
• ATM also applies CRC to the data, with varying
  degrees of error control

42
Summary

• Noise in computer networks
• Error-prevention techniques
• Simple parity and longitudinal parity
  calculations
• Cyclic redundancy checksum
• Stop-and-wait ARQ, Go-back-N ARQ and
  Selective-reject ARQ
• FEC