Transcript EE 3760 chapter 7 - Seattle Pacific University

Data Transmission:
Data and Signals
Based on Chapter 3 of William Stallings, Data and
Computer Communication, 9th Ed.
Kevin Bolding
Electrical Engineering
Seattle Pacific University
Seattle Pacific University
Data Transmission
No. 1
Data Transmission
• Two major aspects of data transmission:
• Data – What you are trying to get to the receiver
• Actual information being sent/received
• Analog (continuous) or digital (discrete)
• Signal – How the data is actually sent
• Electronic or electromagnetic representation of data
• Analog or digital (independent of data type)
Seattle Pacific University
Data Transmission
No. 2
Data
• Data comes in thousands of flavors…
• Audio
• Speech and music are the most common
• Video
• Television, remote monitoring, videos
• Images
• JPEG, GIF, etc.
• Text
• Files, email, text messages
• Various computer formats
• Word documents, Excel documents
• Control information
• Remote operation, commands
Seattle Pacific University
Data Transmission
No. 3
Audio Data
0dB
Power ratio
25dB
-20dB
Telephone
Channel
Speech
70dB
30dB
-40dB
3.1kHz
-60dB
Music
10Hz
100Hz
1kHz
10kHz
100kHz
Frequency
Source: Stallings, Fig. 3.9
Seattle Pacific University
Data Transmission
No. 4
Analog Video (NTSC) Data
1
Portion of TV screen
2
3
4
5
NTSC Television:
480 Lines x 450 pixels
(more or less)
6
7
8
9
10
11
12
13
Interlaced: Odd lines
scanned first, then even
lines
14
15
16
17
18
19
Scan line (even)
Scan line (odd)
Bandwidth lost to horizontal
retrace and vertical flyback
Horiz. Retrace
Vert. Flyback
Source: http://www.ntsc-tv.com
Try: http://www.ntsc-tv.com/images/tv/aa-raster-1.gif for an interesting animation.
Seattle Pacific University
Data Transmission
No. 5
Text and Computer Data
• Text data is human-readable
• Transmitted in the International Reference Alphabet
(IRA), known in the US as ASCII
• Seven/eight bits per character
• Computer data is not human-readable
• May be in any one of thousands of formats (.doc, .xls,
.wav, .mp3, .avi etc.)
• Binary in nature – Interpretation is left to the computer
Seattle Pacific University
Data Transmission
No. 6
Signals
• Signals are the physical representation of data
• Signal must have enough capacity (bandwidth) as the
data being transmitted needs
• Analog signals are continuous in nature
• Contain an infinite number of possible signal levels
• Limited by noise
• Digital signals are discrete in nature
• Finite number of signal levels
• Still limited by noise, but easier to deal with it
Seattle Pacific University
Data Transmission
No. 7
Signal-to-Noise Ratio
• The quality of a signal is judged by how well the
original data can be extracted from it
• Noise will corrupt the signal
Transmitted Signal
Received Signal
• The important measure is the power ratio:
• Received Signal Power/Received Noise Power
• In most cases, the ability to distinguish the signal is
based on the log of the power ratio
Seattle Pacific University
Data Transmission
No. 8
Measuring Signal-to-Noise Ratio
• SNR = Signal Power/Noise Power
• Most signals are observed as a voltage waveform
Typically use Peak Signal
• Power = V2/R
Received Noise
= 1V Average
and Average Noise
5V
4V
3V
2V
1V
0V
Received Signal
= 5V Peak
Received Signal
Both signal and noise
see the same load, R,
so it cancels out
• SNR = (52/R) / (12/R) = 52 / 12 = 25
• In deciBels
• SNRdB = 10 log10(PS/PN) =10 log10(25/1) = 13.97dB
• Note: SNRdB = 10 log10(VS2/VN2) = 10 log10(VS/VN)2 =
20 log10(VS/VN)
• SNR = 20 log10(5/1) = 13.97dB
Seattle Pacific University
x10 if measuring Power,
x20 if measuring Voltage
Data Transmission
No. 9
Telephone Signals
• Speech occupies a band between 100Hz and 7kHz
• Almost all useful information is between 300Hz and
3.4kHz
• Telephone signals (POTS) are electrical
representations of the sound signals
• Bandwidth of 3.1kHz (300 – 3400 Hz)
• S/N ratio of 30dB (Maximum signal power is 1000x the
average noise power)
• S/N ratio (dB) = 10 log10(Signal power/Noise power)
Seattle Pacific University
Data Transmission
No. 10
Video (NTSC) Signals
• An analog signal giving a gray scale value for each pixel
• Synchronizes to the TV’s scanning circuitry, then just
blasted to the screen
• Approximate Analysis:
• Scanning frequency: 525 lines in 1/30 sec. 63.5ms/line, but 11ms
•
•
•
•
•
used for retrace  52.5ms/line
Each line contains approx. 450 pixels
Highest frequency needed when displaying alternating black/white
pattern
• Two pixels per period (high/low portions of wave)
Requires 52.5ms/450 pixels/ 2 pixels/period = 233.3 ns/pixel 
4.2MHz (high end)
Low end: All black or all white  DC (0 Hz)
Bandwidth needed is (4.2 – 0 MHz) = 4.2MHz
Seattle Pacific University
Data Transmission
No. 11
Digital Signals
See http://www.falstad.com/fourier/index.html
for a demonstration of this
• Digital signals are sent as pulses (square waves)
• ‘1’ represented by a high voltage, ‘0’ by a low voltage
• Other representations are possible as well
A square wave:
Requires Infinite bandwidth.
Square wave using finite bandwidth:
Using bandwidth of 6x base frequency
Using bandwidth of 4x base frequency
Source: Stallings, Fig. 3.7
Seattle Pacific University
Data Transmission
No. 12
Data and Signals
Any combination of digital/analog data and digital/analog signals
is possible
Analog Data
Analog Signal
Digital Data
Modem
Analog Signal
Analog Data
Digitizer
Digital Signal
Digital Data
Transceiver
Digital Signal
Seattle Pacific University
Data Transmission
No. 13
Is Digital or Analog “Better”?
• Data is inherently digital or analog
• Digital signals and digital transmission are taking
over
• Better data integrity
• Possible to ensure 100% accurate transmission of
a digital signal
• Better utilization
• Easier to multiplex digital signals
• Security
• Encryption is easy with digital data
Seattle Pacific University
Data Transmission
No. 14