IT101 Introduction to Information Technology

1
IT-101Introduction to Information Technology

• Instructor
• Dr. Peter Pachowicz (ECE Dept.)
• Office hours Wednesday 230-400pm, and by an
  appointment
• Syllabus
• Course description
• Tentative schedule
• Grade
• Exams 85 (Midterm I 20, Midterm II 25,
  Final 40)
• Homeworks 15 (3 homeworks, Note Dont miss the
  deadlines)
• (1 recommended homework not graded)
• Course text

2
Other Issues

• Honor code
• Poll
• Major
• How many IT courses did you take?
• Link for course
• Lecture slides will be posted in PowerPoint
• MOTIVATION
• MY COURSE GOALS

3
Information TechnologyAt least to Webster

• Information (Latin idea, conception)
• Knowledge communicated or received concerning a
  particular fact or circumstance
• Any knowledge gained through communication,
  research, instruction, etc.
• Any data that can be coded from processing by a
  computer or similar device
• Technology (Greek systematic treatment)
• The branch of knowledge that deals with
  industrial art, applied science, engineering

4
Information TechnologyAccording to WhatIs.com

• IT (information technology) is a term that
  encompasses all forms of technology used to
  create, store, exchange, and use information in
  its various forms (business data, voice
  conversations, still images, motion pictures,
  multimedia presentations, and other forms,
  including those not yet conceived). It's a
  convenient term for including both telephony and
  computer technology in the same word. It is the
  technology that is driving what has often been
  called "the information revolution."

5
Components of IT

• Computers
• Telecommunications
• Software
• Creativity
• First IT breakthrough?
• Cave paintings? The written word?
• Greatest IT development?
• Telephone, telegraph, computer, yet to happen?

6
Information System ExampleThe Telephone System
Switching and Signaling System
Transmission Media
Receiver
Transmitter
7
Information System ExampleThe World Wide Web
8
Signals vs. Messages vs. Information

• Signal A physical representation of a process.
  Examples include sound waves, light pulse,
  electronic impulses, etc.
• Message The content of the signal. Examples
  include binary representation, alphanumeric
  characters, etc.
• Information Whatever the content of the message
  conveys to the observer.

9
Digital vs. Analog
DISCRETE
CONTINUOUS
Exists at SPECIFIC POINTS in space and
time Variables can have ONLY SPECIFIC values
Exists for ALL values of space and time Variables
can have ALL POSSIBLE values
10
Analog and Digital Technologies
11
The Natural World is Analog
Human speech is an example of analog
communication. Speech causes air to vibrate with
varying amplitude (volume) and frequency (pitch).
This continuous acoustical waveform can be
detected by a device and converted into an
analogous electrical waveform for transmission
over a circuit.
12
The Computer World is Digital

• Digital computers communicate using 2 discrete
  values. In other words, they speak in binary (0
  and 1).
• Of course, 0s and 1s are not literally
  transmitted
• In an electrical network, variations in voltage
  represent one of the two discrete values.
• In an optical network, pulses of light provide
  the discrete values.
• Recall that the 0s and 1s are the message and
  the pulses of light or voltage variations are the
  signal.
• Two values in different combinations sufficiently
  encode text, numbers, image, and video!
• Note that the telegraph was an early example of
  communications using discrete, electrical pulse
  transmission.

13
The Digital Advantage

• Quality Digital information is resistant to
  signal degradation.
• Copying analog information magnifies noise and
  interference.
• Digital information can be perfectly
  regenerated.
• Security Analog signals can be scrambled
  (through frequency mixing) to provide some
  security. Digital signals can be manipulated
  with complex encryption algorithms for much
  improved security.
• Efficiency Digital information can be easily
  compressed, providing more efficient
  transmission.
• Cost Digital information systems are cheaper
  than analog systems.
• Management Management features can be designed
  into digital signals.

14
The IT Building Block or Digital Native Language

• Binary representation
• 1s and 0s
• What it is physically Presence of or level of
  voltage
• Lets count
• Binary is counting in a base 2 scheme
• Decimal is base 10
• Octal is base 8
• Hexadecimal is base 16

15
Why Use a Code with Only Two Values?

• A binary system is resistant to errors!
• The two symbols are highly distinguishable from
  one another.
• Two values correspond well to the on and off
  states of electronic switches that comprise
  digital computers.
• Simplicity!
• Any Problems with a binary code???

16
How do Digital Systems Produce 0s and 1s?

• Physically, a 1 or a 0 can be produced in several
  ways
• The presence of or level of voltage in an
  electrical network.
• A pulse of light or varying of light intensity in
  an optical network.
• Discrete variations of signal amplitude in a
  radio network like satellite or cellular.
• STORAGE - To store binary data, storage media
  must represent two values.
• Magnetic disk can be magnetized in two directions
  up or down
• Laser disk domains have either a smooth surface
  or pitted surface.
• TRANSMISSION - Two distinct electrical or optical
  quantities are transmitted such as a pulse of
  light and absence of light.
• PROCESSING - Computer circuits can be broken down
  into the fundamental building block, the
  electronic switch (either on or off).

17
Binary Conventions

• Most Significant Bit (MSB) and Least Significant
  Bit (LSB)
• Decimal Example 64
• 6 is the Most Significant Digit
• 4 is the Least Significant Digit
• Binary 1000000
• 1 is the MSB
• 0 on the right is the LSB
• What does 10000002 represent in decimal? One
  method of binary to decimal conversion
• How do I represent 1310 in binary?

18
More Examples

• Convert 1210 to binary
• Convert 1010101 to decimal
• Convert 25510 to binary
• Convert 1110001110 to decimal

19
Thinking Binary

• When dealing with binary, always think in powers
  of 2
• Binary Computer conventions (You need to know
  this!)
• 1 bit can represent two (21) symbols either a 0
  or a 1
• 8 bits is a byte one byte can represent 256 (28)
  symbols
• 1 kilobyte 210 bytes 1024 bytes 8192 bits
• 1 Megabyte 220 bytes 1,048,576 bytes
• How many bits are necessary to encode the
  uppercase alphabet? Upper and lower case?

20
Binary Multipliers

• Kilo (K) 210 1,024
• Mega (M) 220 1,048,576
• Giga (G) 230 1,073,741,824
• Tera (T) 240 1,099,511,627,776
• Peta (P) 250 1,125,899,906,842,624
• Exa (E) 260 1,152,921,504,606,846,976
• Zetta (Z) 270 1,180,591,620,717,411,303,424
• Yotta (Y) 280 1,208,925,819,614,629,174,706,1
  76
• Note You use these factors with base 2 numbers,
  not base 10 numbers!

21
Something to Remember

• Bits are often used in terms of a data rate, or
  speed of information flow
• 56 Kilobit per second modem (56 Kbps)
• A T-1 is 1.544 Megabits per second (1.544 Mbps or
  1544 Kbps)
• Bytes are often used in terms of storage or
  capacity--computer memories are organized in
  terms of 8 bits
• 256 Megabyte (MB) RAM
• 40 Gigabyte (GB) Hard disk

22
Lets Be Clear! Bits v. Bytes

• 56Kbps represents 56,000 bits per second (56,000
  is a base 10 number)
• 56KB represents 56 x 210 bytes (bytes--8
  bits--base 2 number)
• or 56 x 1024 bytes 57,344 bytes
• or 57,344 bytes x 8 bits 458,752 bits

23
Practical Use

• Everyday stuff that uses 2X
• 32-bit sound card
• 64-bit Video Accelerator card
• 128-bit encryption in your browser
• Is 64-bit twice as good as 32-bit?
• 32 bit 232 4,294,967,296 bits
• 64 bit 264 1.8 x 1019 bits
• 128 bit 2128 3.4 x 1038 bits

24
Other Ways to Count

• Octal--base 8
• 0, 1, 2, 3, 4, 5, 6, 7
• 810 108
• Represents 3 bit code see page 55 of text
• Hexadecimal--base 16
• 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, A, B, C, D, E, F
• Represents 4 bit code see page 55 of text
• Compresses notation F16 178 11112 1510
• Note on page 56 of text that 7-bits are being
  converted into hexthe authors inserted a
  leading 0 into each group of 7 bits to make a
  byte
• Notice that if you break each byte into a 4 bit
  word you can easily convert binary into hex

25
Fixed Length Codes

• Fixed length codes represent information in a
  format someone has agreed to
• American Standard Code for Information Exchange
  (ASCII)--see Appendix A in text
• Structure is a 7 bit code (plus a parity bit or
  an extended bit in some implementations
• ASCII can represent 128 symbols (27 symbols)
• INFT 101 is 73 78 70 84 32 49 48 49 (decimal)
  or
• 49 4E 46 54 20 31 30 31 (hex)
• 1001001 1001110 1000110 1010100 0100000 0110001
  0110000 0110001 (using Appendix A)
• How do you spell Lecture in ASCII?

26
Extended ASCII Chart

• You can easily convert hex to binary
• Upper case D is 4416
• 416 is 01002
• Upper case D is then 0100 0100 in binary
• Note 8 bits used

27
Extended ASCII Continued

• Another example You want to represent the Yen
  sign ()
• You see it is 9D
• 916 910 10012
• D16 1310 11012
• The sign in binary is 1001 1101

28
Other Text Codes

• Extended Binary Coded Decimal Interchange Code
  (EBCDIC) used by IBM-- 8 bit (28 bits) 256
  symbols
• Unicode is 16 bit (216) 65,536 symbols
• World Wide Web supports many languages
• Unicode supports Latin, Russian, Cherokee, other
  alphabet representation
• www.unicode.org

29
Error Detection and Correction

• Errors in transmission can be caused by many
  factors
• Clever code construction or additional
  information added to the transmission can
  increase the odds of the sent information being
  received correctly

30
Parity Bit

• An extra bit is appended to the code
• Even parity is when the bit is set so that the
  total number of 1s in the word is even
• 11 ? 11 0
• 10 ? 10 1
• Odd parity is when the bit is set so that the
  total number of 1s in the word is odd
• 11 ? 11 1
• 10 ? 10 0

31
Parity Errors

• Even parity is set 10 0 is received.
• Error present, but dont know which one is the
  bit in error
• Parity can detect errors, but cant correct them
• Odd parity is set 1111011 0 is received.
  Error?
• Even parity is set 1111011 0 is received.
  Error?

32
Repetition

• Repetition is the provision of extra bits to
  ensure the information is received correctly
• Original data 1 0 0 1
• Transmitted 111 000 000 111
• Received 011 000 001 111
• Errors in the first and third bits detected
• Errors in the first and third bits can be
  corrected

33
Redundancy Check

• Symbols are given a parity bit
• Total word is given a redundancy check
• The first bit of each symbol in the word is given
  parity, then the second bit of each symbol in the
  word is given parity
• The additional parity symbol is given its own
  parity and then appended to the transmitted
  information

34
Redundancy Check

• Information to be sent 00 01 10 11
• With even parity becomes 00 0 01 1 10 1 11 0
• First bits are 0 0 1 1 Odd parity bit 1
• Second bits are 0 1 0 1 Odd parity bit 1
• Odd parity bits 1 1 Even parity bit 0
• Transmitted 000 011 101 110 110

35
Redundancy Check Error

• Transmitted 000 011 101 110 110
• Received 000 111 101 110 110
• Even parity tells us that the second symbol has
  an error
• Comparing Odd parity with the first bit in each
  symbol shows us that the first bit in the second
  symbol should be a 0
• Comparing Odd parity with the second bit in each
  symbol shows us that everything is OK

36
Functions

• In mathematics, a function is an association
  between two sets of values in which each element
  of one set has one assigned element in the other
  set so that any element selected becomes the
  independent variable and its associated element
  is the dependent variable. Thus, in
• y f(x)
• y is said to be a function of x.

37
Functions

• Common functions we might deal with
• Functions of Time
• Functions of Space
• Functions of Frequency
• Functions of Power
• Usually, functions are represented as a
  relationship on an x-y axis graph

38
The Sine Wave is a Function
39
Continuous v. Discrete Functions

• Continuous Functions are just that over the
  function, there is an infinite number of points
  making up that function. Or between any two
  points you can pick another one
• Discrete Functions A discrete function has a
  finite number of elements (or values)
• Lets draw the difference!

40
Continuous v. Discrete

• In the real world many phenomena can be
  characterized as continuous (analog)
• Mercury thermometer
• Dial gauges
• Sound
• Perceived sight
• In the digital world phenomena are coded in
  terms of 0s and 1sexplicit values
• They are ____________ (fill in the blank)

41
Temperature

• Mercury thermometer reflects a temperature that
  can continuously vary over its range of
  measurement
• Digital thermometer would require an infinite
  number of bits to accomplish the same thing
• So if we are building a digital thermometer, we
  must make some choices
• Precision (number of bits we will use) v. cost of
  chip
• Accuracy (how true is our measurement against a
  standard)

42
Temperature Waveforms
43
Thermometer Engineers

• What is the range we wish to measure?
• How many bits are we willing to implement?
• We want to measure -40º F to 140º F
• Total measurement range 180 º F
• If our thermometer measures in 1º increments, we
  need to represent 180 steps
• We can do this with an 8 bit code (256 values)
• Accuracy issue at what temperature does the
  thermometer increment or decrement? And is it the
  right temperature? (Not dependent on the coding
  scheme we choose!)

44
Design Review

• CEO wants the thermometer to read to the
  hundredth of a degree
• How many bits are required?
• 180 x 100 18000 steps
• Can implement with a 15 bit code (215 32768)
• Note wasted bits
• Precision now to the hundredth degree
• Is a precise instrument the same as an accurate
  one?
• How do we build the code now?

45
Thermometer Coding (One Answer)
46
Thermometer Coding (Another Way)
47
What Was That?

• 2s complement notation is a method to represent
  negative numbers
• Take the positive binary representation with the
  MSB as a sign bit (a 0 makes the representation
  positive)
• 01001 (910)
• Take the binary complement (flip the bit values)
• 10110
• Add a binary 1
• 10110 1 10111 (-910)
• Ill specifically tell you to do something in 2s
  complement on a homework or test

48
Speaking Digital

• Digital cant absolutely capture analog
  (continuous) functions
• Trades are made in length of code-word
• CDs use 16 bits
• So why is digital better?
• Accurate reproduction
• Error detection and correction
• 0s and 1s are easier for electronic systems to
  deal with (encryption, multiplexing,
  regenerating, amplifying)

49
Audio Waveform
50
Storage Needed

• Digital audio signal 16bit per sample
• Sampling every 0.0000227 sec.
• It means, we need total
• 1,411,200 bits for every second of music
• 16 bit per sample
• 44,100 samples per sec.
• 2 channels (stereo signal)
• Added error correction will multiply this storage
  requirement

51
Protocols

• Believe it or not, you use protocols every day
• When you see a red, octagonal sign, you_____
• When you pick up the telephone when it rings, you
  say _______
• The procedure in checking out a library book
• Using the ATM
• Protocols give structure and provide rules, but
  they arent based on anything but human
  convention, agreement and understanding

52
Protocols

• IT is built on protocols
• Hypertext Transfer Protocol (HTTP)
• Simple Mail Transfer Protocol (SMTP)
• Transmission Control Protocol/Internet Protocol
  (TCP/IP)
• And just about everything else we will talk about
• Weve already discussed the convention of the
  byte
• Four bits are called nibbles
• Four (or eight) bytes is a word

53
Who Makes Protocols?

• International Organizations
• International Telecommunications Union (ITU)
• International Standards Organization (ISO)
• National Organizations
• American National Standards Institute (ANSI)
• Professional Organizations
• Institute of Electrical and Electronic Engineers
  (IEEE)
• Trade Groups/Consortiums Sony/Phillips and the
  CD
• Companies Microsoft, Cisco, Bell (back in the
  day), 3Com, others..

54
Why Protocols?

• Protocols provide the rules under which a desired
  function can occur
• Greatly simplify production and development of
  new products compatibility with other vendors
  and older equipment
• Bring order to chaos
• Text has example of 140 different formats of
  floppy disk storage on the same 5 1/4 inch media
• Provide new and better functionality

55
Protocol Problems

• Computers arent all that capable of independent
  thought
• New protocols drive obsolescence
• Microsoft XP wont use the 95 kernel (allegedly)
• And perhaps obsolescence of the competing
  protocol (see next bullet)
• The best protocol doesnt always win
• Market forces and inertia sometimes are the
  victor (DVD v. DiVX)
• Whats the standard?
• Physical devices 3.5 floppy disk contains
  160tracks 18 sectors 512B 1.44MB
• In a word Interoperability

56
Some Examples (Data Transmission)

• How do we know where the bit stream begins?
• Start bits stop bits (Guess what? These are
  protocols!)
• Flags--used in Ethernet, High-Level Data Link
  Control (HDLC)

Flag 01111110
57
Protocol Take-a-Ways

• Check the protocols you are using
• Check them again
• Make sure the other guy is using the same
  protocols
• Check again
• New protocols build on older ones to support
  interoperability and are approved by standards
  bodies to ensure availability and compliance
• But specific vendor implementation may cause
  interoperability problems (thats why you check)
• Some protocols are introduced as bridges between
  incompatible protocols -- Rich Text Format (RTF)
  protocol

58
Visual Representation and Display

• A picture is worth ten thousand words
• But it takes a whole lot more than 10,000 bits
  usually!
• A visual image is a powerful way to convey
  information
• The need to process a huge amount of image/video
  data requires a lot of computing power
• Specialized processors
• New type of computers
• A very dynamic growth in consumer electronics

59
Image Issues

• The world we live in is analog
• Another way to look at it--continuously variable
• Problem for digitizing, or making discrete
• But
• We are producing information for human use
• Human vision has limitations
• Take advantage of this fact
• Produce displays that are good enough

60
Digital Info for Humans

• Many digital systems take advantage of human
  limitations (visual, aural, etc)
• Human gray scale acuity is 2 of full brightness
• Or 50 gray levels can capture what most folks
  can detect
• Human eye can resolve about 60 lines per degree
  of visual arc
• 8.5 x 11 sheet of paper at 1 foot (landscape)
  contains 49.25 degrees horizontal and 39 degrees
  vertical
• 49.25 degrees x 60 2955 horizontal lines we can
  just distinguish
• 39 degrees x 60 2340 vertical lines we can just
  distinguish
• These numbers give us a clue about how long our
  code will need to be to capture images

61
Cameras and Image Formation

• Basic image formation process
• Essential components
• Object/scene
• Lens
• Image recording medium
• Display device may correct the inverted condition
• Projection 3D to 2D

62
Other Types of Image Formation

• Radar, Sonar, X-rays, CAT scan
• Differ from traditional camera
• Type of energy used to form image
• Geometry of the system that relates the location
  of the objects in the real world to the image
  world

63
Radar

• Radar Radio Detection and Ranging
• TV weather forecasts
• Sends a narrow beam of radio waves in a
  particular direction and waits for reception of
  some reflected energy
• Distance and angle of transmission (polar
  coordinate system)

64
Pixels and Lines

• A pixel is the smallest unit of representation
  for visual information
• In order to form a black line, you need two rows
  of pixels (one black and one white) to give a
  visual clue of transition
• So two rows of pixels are needed per line
• Go back to our paper example
• Number of pixels needed to represent total image
• 2(2955 horizontal) x 2(2340 vertical)
  27,658,800 pixels

65
Check for Yourself

• On the CD that accompanies the text, play with
  the visual representation applets
• Some changes to the level of pixels or colors
  affect very little your ability to discern the
  image
• After a time degradation becomes very apparent
• On pages 84 and 85 you can see the effects of
  going from a 6-bit to 1-bit grayscale code

66
Picking the Codes

• The assignment of a code to a particular analog
  value is called Quantization
• Quanitization results in the loss of some
  information and precision
• You are taking a continuous function and
  assigning discrete values, resulting in some
  values not being exactly represented
• More detail on Quanitzation in a later lecture

67
Spatial Resolution

• Image represented in a finite number of small
  picture elements
• Represent a single intensity (brightness) level
• Usually arranged in rectangular grid
• Tradeoff
• Why do we want more pixels?
• Why do we want fewer pixels?

13 x 13 grid 169 pixels
68
How Many Pixels Should We Use?

• If too few pixels used, image appears coarse

16 x16 (256 pixels)
64 x 64 (4096 pixels)
69
Digitized Image

• Process of breaking a continuous image into a
  grid of pixels is called pixelization, sampling,
  scanning or spatial quantization.

70
Shades of Gray

• Each pixel must be converted to binary data.
• Suppose 8 bits are used to represent the
  brightness at each pixel.
• How many brightness levels?
• 28 256 brightness levels (black gt white)
• Quantization process of rounding off continuous
  values so they can be represented by a fixed
  number of binary digits.

71
Shades of Gray
A 6-bit (64 gray levels) image

• A 3-bit (8 gray levels)
• image

A 1-bit (black and white only) image
72
How Much Storage Is Needed?

• Total number of bits required for storage total
  number of pixels number of bits used per pixel
• For example Black and white photo
• 64 x 64 pixels
• Use 32 gray levels (5 bits)
• 64 x 64 x 5 20,480 bits 2560/1024 bytes
  2.5KB
• Remember data storage is in bytes
• KB represents 210 or 1024 bytes

73
Color Representation

• Instead of grayscale, assign coding to the levels
  of each of three colors--Red, Green, Blue (RGB)
• Another approach is Hue (light spectrum),
  Luminance (brightness) and Saturation (dilution
  by white) (HLS)
• Both systems use 3 bits per component (RGBHL or
  S), for a total of 9 bits--512 different levels
  of representation
• Limited color scheme, so additional palettes are
  available, each with a 512 level representation
  (Color Palette Representation)
• Windows supports 32-bit color--much larger choice
  of colors than with RGB!

74
Video

• Human perception of movement is slow
• Studies show that humans can only take in 20
  different images per second
• We can detect higher rates of flicker, but only
  to about 80 cycles per second
• Movies show 24 frames per second, but flash at 48
  per second
• TV works similarly, but instead of a frame, TV
  refreshes in lines across the tube
• Whats going on? Make discrete a continuous
  function--it is possible to capture the image
  with bits!

75
Your Computer Display

• Described in terms of pixels
• 640 x 480
• 800 x 600
• 1024 x 768
• At 60 Hz, how many pixels are being refreshed in
  a 1024 x 768 resolution per second? 1024 x 768 x
  60 47,185,920
• Encoding Supports some number of colors
• 256 (one byte)
• 65536 (two bytes)
• 16,777,216 (three bytes or indexed)--see p 99 of
  text
• Using number of pixels computed above, how many
  bytes are being processed per second for 65,536
  colors? 47,185,920 x 2 bytes 94,371,840 bytes
  per second

76
Compression

• Compression techniques can significantly reduce
  the bandwidth and memory required for sending,
  receiving, and storing data.
• Most computers are equipped with modems that
  compress or decompress all information leaving or
  entering via the phone line.
• With a mutually recognized system (e.g. WinZip)
  the amount of data can be significantly
  diminished.
• Examples of compression techniques well discuss
  today
• Compressing BINARY DATA STREAMS
• Variable length coding (e.g. Huffman coding)
• Universal Coding (e.g. WinZip)
• IMAGE-SPECIFIC COMPRESSION
• GIF and JPEG
• VIDEO COMPRESSION
• MPEG

77
Variable Length Coding (Compression)

• ASCII is a fixed length code--8 bits for Extended
• Certain letters are used more frequently than
  others in the English language
• This fact is the basis for Morse Code
• . E - T --.. Z
  --.- Q
• Why bother?
• Saves transmission time, reduces storage
  requirements

78
Definitions

• Message Values or meanings to be conveyed
• Data A stream of symbols (voltages, 1s/0s)
• Information Amount of surprise in the message
• Take pi can be symbolized by p or a series of
  non-repeating digits 3.141592654.
• Easier to send p than the whole string of digits
• We already know (a priori) what ps value is
• We can take advantage of this fact and compress
  the data we send

79
Basic Information Theory

• Claude E. Shannon developed modern Information
  Theory at Bell Labs in 1948
• He saw the relationship between the probability
  of appearance of a transmitted signal and its
  information content
• This was a huge realization that allowed for
  compression techniques

80
Probability

• Shannon found that information can be related to
  probability (Others had done work as well, but
  Shannon gets credit for Info Theory)
• Fair coin toss Heads or Tails each has a
  probability of .50
• In two tosses, both heads probability is .5 x .5
  .25
• In three tosses, all tails .5 x .5 x .5 .125
• We compute probability this way because the
  result of each toss is independent of the results
  of other tosses

81
Entropy

• If the probability of a binary event is .5 (like
  a fair coin), then, on average, you need one bit
  to represent the result of this event.
• As the probability of a binary event increases or
  decreases, the number of bits you need to, on
  average, represent the result decreases
• The figure is expressing that unless an event is
  totally random, you can convey the information of
  the event in fewer bits, on average, than it
  might first appear
• Lets step through the store example in the text

82
How do you measure code effectiveness?

• The authors describe the store example
• 80 of the customers are male, 20 female
• For now, just accept the codes assigned
• M-M 0 M-F 10 F-M 110 F-F 111
• (1)(.64) (2)(.16) (3)(.16) (3)(.04) 1.56
  bits/event
• The leading number is the number of bits
  contained in the code
• The fractional number is the probability of that
  codes occurrence
• We achieved 0.722 lt 1.56 lt 2 bits/event
• Lower bound and upper bound concept

83
Huffman Coding

• You need to understand the probability of event
  occurrence
• Then you order the events in probability of
  occurrence
• Then you group the events in pairs of events with
  lowest probabilities
• Compute the probability of occurrence for the
  pair (you add the probabilities)
• Keep pairing, even if you pair previous pairings
  (keep pairing lowest unpaired probabilities!)
• Add your codes to the events--watch that your
  codes are shortest for the most common
  occurrences and that each code can be uniquely
  identified

84
Example
A (.35)
B (.25)
C (.13)
D (.12)
E (.1)
F (.04)
G (.01)
85
Huffman Coding

• Using our just constructed code, how should you
  decode this string?
• 0000111010110001000000111110011011000
• 00 00 1110 101 100 01 00 00 00 1111 100 110 110
  00
• A A F D C B A A A G C
  E E A
• How do we know this is the code? Every code is
  unique in form--note the wasted bit characters
• That is there is no such thing as 0 or 11
  etc...

86
Universal Coding

• Huffman has its limits
• You must know a priori the probability of the
  characters or symbols you are encoding
• Universal coding allows you to compress as you go
• Lempel-Ziv coding one form of Universal Coding
  presented in the text
• Used by WinZip to compress info

87
Variable Codes

• Takes advantage of probabilistic nature of
  information
• If we had total uncertainty about the information
  we are conveying, fixed length codes are better
• Variable codes achieve their efficiency over
  longer information strings
• Lempel-Ziv achieves large efficiencies over long
  strings of info
• Over short strings, Lempel-Ziv is very
  inefficient compared to Fixed or Huffman

88
Introduction Computer Architectures and Software

• Read Chapter 5 in Optional text for more detail
  (2 copies are on reserve in the library)
• The IT world is based on Computers--you should
  have a basic understanding of how the computer
  works and about the software that drives it

89
ENIAC

• What were computers like just over 50 years ago?
• 1946 - ENIAC
• Used plugboards and switches to program
• Occupied 1,000 square feet
• Used vacuum tubes
• Developed at UPenn
• Funded by U.S. government

Electronic Numerical Integrator and Computer
90
Computer Technology Evolution
Ultra Large Scale Integration gt 100 million
devices per chip
Very Large Scale Integration Post 1978 100,000 -
100 million devices per chip
Vacuum Tubes 1946-1957
Transistors 1958-1964
Large Scale Integration 1971-77 3000- 100,000
devices per chip
Medium Scale Integration Pre-1971 100-3000
devices per chip
Small Scale Integration 1960s Up to 100 devices
per chip
INTEGRATED CIRCUIT
91
Moors Laws

• Intel pioneer, Gordon Moore, predicted in 1965
  that the number of transistors on a chip would
  double every 18 months.

92
The Desktop PC
Storage
Central Processing Unit
Input/Output
Memory
93
Input/Output

• Two ways to think of I/O
• 1 At the board/chip level (Physical Connections)
• 2 What is connected to the board (User I/Os)
• Depends if you are a computer architect or a
  peripheral device designer/user
• The board
• Parallel, serial, network, game, digital ports
• What is connected to the board
• Printer, scanner, keyboard, joystick, modem,
  speaker, monitor, network card

94
Input/Output
?
Sound Board
?
Game Board
CPU
?Monitor
Graphics Board
?
Serial Port

Parallel Port
?Mode
Serial Port
Local Area Network
Network Port
?
ADC
Analog signal source
input/output
bus
95
Storage

• Used by the computer to store programs,
  information, data (Write)
• Used by the computer to access information and
  programs when necessary (Read)
• Some storage can do both read and write, some
  only one
• Some storage can randomly access information,
  some can only sequentially access
• Magnetic
• Hard disk, floppy disk, ZIP disk, tape
• Optical
• CD, DVD
• Solid State
• Flash cards, PCMCIA cards

96
Memory

• Memory helps computers operate faster and more
  efficiently
• Random Access Memory (RAM) --- Temporary
• Memory chips in a PC
• Virtual Memory
• Cache Memory (on the CPU)
• Read-Only Memory (ROM) --- Long-Term
• CD-ROM
• DVD-ROM

97
The Central Processing Unit (CPU)

• The brains of the computer
• Performs calculations and completes instructions
• Performance based on clock speed--how many
  instructions can it accomplish in a cycle
• Pentium 4 --- 2.0 GHz chip operates at 2 billion
  cycles per secondcan do one instruction per
  cycle per pipeline

98
The CPU
Storage
Memory
Arithmetic Logic Unit (ALU)
Control Unit
Registers
Cache Memory
Input/Output
Flags
99
Four Stages of CPU Operations

• Fetch - Seeks instructions from outside source
• Decode - Analyzes the instructions to determine
  which of the chips circuits should be used for
  processing
• Execute - Performs the actual instructions
• Store - Places processing result in appropriate
  place

Comparing a Dime to a Microprocessor
100
Simple Instruction

• You want your computer to add 1 2 4
• MOV 1, R0 (Move the number 1 into Register 0)
• MOV 2, R1 (Move the number 2 into Register 1)
• ADD R0, R1 (Add R0 to R1 and put the result in
  R1)
• MOV 4, R0 (Move the number 4 into Register 1)
• ADD R0, R1 (Add R0 to R1 and put the result into
  R1)
• Move, Add are examples of instruction sets
• Real computers have many more Registers

101
Chip Types

• Complex Instruction Set Computer (CISC)
• Many forms of instructions (some special purpose,
  but chip must support them all)
• x86, Pentium
• Reduced Instruction Set Computer (RISC)
• Specific list of supported instructions (Want to
  do something else? Find a combination of
  instructions to accomplish the task)
• Power PC (Motorola), Alpha, IBM RISC System/6000,
  Sun SPARC, MIPS
• Difference is speed of execution RISC is faster,
  but may require longer instruction combinations
  to accomplish the same task as a CISC

102
Hierarchy of Software(Above the Abstraction
Level i.e. 0s and 1s)
Application
Computer Language (High Level Language)
Operating System
Assembly Code
Machine Language
103
Machine Language

• Machine Language
• 01100100100101010
• Lowest level language
• Consists of elementary instructions directly
  recognized by the CPU
• Provides numerical codes directly recognized by
  the CPU
• Machine language programming produces a string of
  numbers
• Cannot be used by programmers any longer

104
Assembly Code

• Assembly Code Most basic of abstraction
• ADD R1, R2
• MOV ADDR, R1
• Different computer systems (RISC v. CISC) use
  different languages
• Breaks down instructions to their most basic--at
  the level of implementation of the chip (or what
  instructions a particular chip will understand)

105
Operating System

• OS provides for common system tasks
• Display information on a monitor
• Print services
• User interface
• Text based
• Graphical User Interface (GUI)
• MS/DOS, Windows 3.0, Windows 95/98/2000/Me/XP
• Microsoft NT/2000 Professional/Server/XP
  Professional
• Mac OS X (and other earlier)
• Linux
• HP Unix
• Sun Solaris
• IBM OS/2

106
Examples of Operating System
MS-DOS Introduced in 1981 Microsofts first
O/S Microsoft Disk Operating System (MS-DOS) Text
based O/S -- C/gt
Mac OS Appeared in 1984 Apple Macintosh Icons
and Graphical User Interface (GUI)
IBM OS/2 Roughly 1992 Split with Microsoft Never
took off
Microsoft Windows Dominates PC market Windows 3.x
in 1990 Windows 95 and 98 Windows NT Windows
2000 Windows XP
Unix Variations IBMs AIX Hewlett Packards
HP/UX Suns Solaris Linux Others
107
Computer Language(High Level Language)

• Used to program instructions for computer use
• Eases programming, since the Compiler
  translates almost human syntax into lower level
  code
• Java
• Fortran
• C
• Visual Basic
• Pascal
• Ada
• HTML

108
Programming Languages
Visual Basic Microsofts visual
language Provides an array of tools that
decrease development time
Basic (Dartmouth College, Kemeny and Kurz) Simple
language students could learn 1964
C and C Developed originally as C/Unix
in 1974 C is object oriented version
FORTRAN (Formula Translator) Developed by
IBM Science/engineering 1957
Pascal Once popular with serious
programmers 1970s Appeal has diminished
COBOL (Common Business Oriented Language) Pushed
by U.S. Govt. 1960
Java Developed by Sun O/S independent
HTML HyperText Markup Language describes
documents on the Web
XML eXtensible Markup Language More powerful
successor to HTML
109
Application Software

• This is how we usually interface with the
  computer
• MS Word
• MS PowerPoint
• Lotus Notes
• Quicken TurboTax
• AOL Instant Messenger
• Etc., etc., etc..