Long Distance Digital Connection
Download
Report
Transcript Long Distance Digital Connection
Long Distance Digital
Connection
CSIT 220 (Blum)
1
Telephony
• The prefix “tele” refers to doing or moving
something over a distance
• The root “phone” refers to sound, especially that
connected with speech
• Thus the telephone is a device that transmits
speech over long distance
• Telephony (te lef’ e nē) is the science/technology
associated with the transmission of speech over
long distances
CSIT 220 (Blum)
2
Telephony (Cont.)
• Because an extensive phone system was
already in place, long distance computer
networks could use the phone system to
transmit data and avoid the expense of lying
down new transmission cables.
• The divisions between telephony, telecommunications and computer networking
have become blurred.
CSIT 220 (Blum)
3
Analog or Digital
• Because speech is itself an analog signal
(pressure waves in the atmosphere),
traditional telephony has been based on
transmission of analog electric and/or
electromagnetic signals.
• However, the idea of digitizing audio
signals predates computer networks by
decades.
CSIT 220 (Blum)
4
PCM
• Shannon and co-workers (Oliver and Pierce)
proposed Pulse Code Modulation in 1948.
• Recall that when a digital signal reaches a
repeater, small amounts of noise can be eliminated
from the signal before it is amplified and
forwarded; whereas when an analog signal reaches
a repeater the noise is amplified right along with
the information.
• In an analog signal, all values are valid and so one
cannot distinguish between the noise and the
signal.
CSIT 220 (Blum)
5
Digitizing
• Consider for example an analog voltage
signal. It can be continuous in two senses:
1. The voltage varies continuously in time
2. At a given instance, the voltage can take on
any value from a continuum.
• To digitize the signal, the time continuum
and the voltage continuum have to be
converted into discrete sets of values.
CSIT 220 (Blum)
6
Analogy: Digitizing an image
Discretize color
CSIT 220 (Blum)
Discretize space
7
Sampling
• Breaking up the time continuum is known as
“sampling.”
• Motion pictures are an example of sampling: a
rapid succession of snapshots (still pictures) are
taken, if the sampling frequency (the number of
pictures/frames per second) is sufficiently high,
the brain perceives the playback as continuous
motion.
• Muybridge demo
CSIT 220 (Blum)
8
Continuous values
(pseudo)-Analog wave
1.5
1
0.5
0
-0.5
-1
-1.5
CSIT 220 (Blum)
Continuous in time
9
Sampled Wave
1.5
1
0.5
0
-0.5
-1
-1.5
CSIT 220 (Blum)
10
Another Nyquist Theorem
• Recall that signals can be thought of as
being comprised of sine waves of various
frequencies (Fourier).
• Nyquist says that to accurately represent a
signal, one’s sampling frequency must be at
least double its highest constituent
frequency.
– One wants to sample its ups and its downs.
CSIT 220 (Blum)
11
Nyquist Sampling Example
• In the following sequence of graphs, a sine wave
is sampled.
• The frequency of the sine wave is doubled each
time, while the sampling frequency is kept fixed.
– Case E does not resemble a sine wave but alternates up
and down with the correct frequency – i.e. the sampling
has captured the most essential feature
– Case F oscillates very quickly (alternating up and
down), but its amplitude seems to vary at a much lower
frequency. This was not a feature of the actual wave
being sampled.
– Case G only has the slowly varying feature whereas the
actual wave sampled varying quite rapidly.
CSIT 220 (Blum)
12
A: sf=10, f=0.159
sf: sampling freq. F: freq.
1.5
1
0.5
0
-0.5
-1
-1.5
The sampling frequency is much higher the “actual” frequency
and the behavior is captured well.
CSIT 220 (Blum)
13
B : sf=10, f=0.318
1.5
1
0.5
0
-0.5
-1
-1.5
CSIT 220 (Blum)
14
C : sf=10, f=0.637
1.5
1
0.5
0
-0.5
-1
-1.5
CSIT 220 (Blum)
15
D : sf=10, f=1.273
1.5
1
0.5
0
-0.5
-1
-1.5
CSIT 220 (Blum)
16
E : sf=10, f=2.546
1.5
1
0.5
0
-0.5
-1
-1.5
The sampling frequency is about 4 times higher than the actual
frequency – the actual oscillation is captured but the shape is not.
CSIT 220 (Blum)
17
F : sf=10, f=5.093
1.5
1
0.5
0
-0.5
-1
-1.5
The sampling frequency is about 2 times higher than the actual
frequency – the actual oscillation is captured, but the shape is not
and spurious low frequency effects have been introduced.
CSIT 220 (Blum)
18
G : sf=10, f=10.186
1.5
1
0.5
0
-0.5
-1
-1.5
The sampling frequency is about equal to the actual frequency – the actual
oscillation is missed entirely, only the spurious low frequency effects remain.
It’s like we sampled Mariah Carey and we got Barry White.
CSIT 220 (Blum)
19
The 8000 Hz decision
• Most humans can hear sounds ranging in
frequency from 20 Hz to 20,000 Hz.
• But for purposes of digitizing and transmitting
speech, the telephone industry decided that they
wanted to represent accurately signals in a range
up to 4000 Hz.
• Thus they opted for a sampling frequency of 8000
Hz (a la Nyquist).
– A bandwidth of 3000 Hz was considered sufficient, but
it was upped to 4000Hz.
CSIT 220 (Blum)
20
Period
• Whereas the frequency (f) is the number of cycles
that go by in a set amount of time (usually a
second), the period (T) is the amount of time
required for a single cycle.
• The frequency and period are reciprocals
f = (1/T) or T=(1/f)
• If the sampling frequency is 8000 Hz, the
sampling period is 0.000125 s = 125 s
(microseconds) = (1/8000 Hz).
CSIT 220 (Blum)
21
The other half of the problem
• At the instant one is sampling, the signal can still
take on an infinite number of values.
• Digitizing requires one to choose a discrete set of
allowed values.
– For example, to digitize an image one can choose two
values (black and white) or allow for shades of gray or
allow for combinations of red, blue and green, etc.
• For the phone system, it was decided that 256
values would be allowed.
– 256 values can be represented by 8 bits.
CSIT 220 (Blum)
22
Sine: 5 values
1.5
1
0.5
0
-0.5
-1
-1.5
CSIT 220 (Blum)
23
Sine: 9 values
1.5
1
0.5
0
-0.5
-1
-1.5
CSIT 220 (Blum)
24
Sine: 17 values
1.5
1
0.5
0
-0.5
-1
-1.5
CSIT 220 (Blum)
25
Sine: 33 values
1.5
1
0.5
0
-0.5
-1
-1.5
CSIT 220 (Blum)
26
A digital audio channel
• So the telephone company’s standard for
digital audio required
64,000 bits/second
= 8,000 samples/second 8 bits/sample
• This 64 Kbps rate can be seen throughout
telephony and is known as DS0 (digital
signal).
CSIT 220 (Blum)
27
T-carrier system
• Bell introduced the first successful system
for digitized voice transmission using the
64 Kbps rate (DS0) in the 1960s.
• The rate has been kept as a standard and the
basis for subsequent standards.
• The US and Europe developed separate
standards.
• The US standards were set up by ANSI
(American National Standards Institute).
CSIT 220 (Blum)
28
DS-X Standards
DS0
Bit Rate
64 Kbps
Mult.
1
T (US)
——
E (Eur.)
——
DS1
1.544 Mbps
24
T1
——
2.048 Mbps
32
——
E1
DS1C
3.152 Mbps
48
——
——
DS2
6.312 Mbps
96
T2
——
——
8.448 Mbps
128
——
E2
C: concatenated
CSIT 220 (Blum)
29
DS-X Standards (Cont.)
—
Bit Rate
34.368 Mbps
Mult.
512
T (US)
—
E (Eur.)
E3
DS3
44.736 Mbps
672
T3
—
—
139.264 Mbps 2048
—
E4
DS4/NA 139.264 Mbps 2176
—
—
DS4
274.176 Mbps 4032
—
—
—
565.148 Mbps 8192
—
E5
CSIT 220 (Blum)
30
T-carrier system
• A T-1 line (1.544 Mbps), commonly used by
businesses to connect to their Internet service
provider (ISP), corresponds to 24 DS0 channels.
• A T-3 line (44.736 Mbps) corresponds to 28 T-1
lines or 672 (2428) DS0 and is commonly used
by ISPs.
• One can also lease “a fractional T-1,” in which one
rents some portion of the 24 channels in a T-1 line,
with the other channels going unused.
CSIT 220 (Blum)
31
A leased line
• Unlike dial-up connections, a leased line is always
active. It is a permanent telephone connection
between two nodes.
• Usually used by businesses to connect distant
offices.
• Typically one pays a fixed monthly rate based on
the distance between the nodes and the speed of
the circuit.
• The line is used exclusively by the lessee, so the
carrier can assure a given level of quality.
CSIT 220 (Blum)
32
T1 and T3
• T1 and T3 lines are entirely digital.
• They use pulse code modulation (PCM) and TimeDivision Multiplexing (TDM).
• They provide full duplex capability by using four
wires
– two wires (one for signal and one for return) for
receiving
– two for sending
• Originally the four wires were two twisted pair
wires, but now they can be coaxial cable, optical
fiber, or even wireless media like digital
microwave.
CSIT 220 (Blum)
33
T1 for Voice (Shay Fig 3.20)
CSIT 220 (Blum)
34
T1 Frame
• T1 can be used for voice or for data transmission.
• Voice signals are sampled at 8000 Hz and each
sample is encoded using 8 bits.
• With 24 such channels being multiplexed (TDM),
a 192-bit frame (24 channels 8 bits/channel) is
sent every 125s.
• One bit separates consecutive frames, so each
frame is actually a 193-bit block.
• The 193 bits/frame multiplied by 8,000 frames/sec
yield 1.544 Mbps data rate.
CSIT 220 (Blum)
35
T1 (Cont.)
• T1 lines typically use copper wire and within or
even between cities (metropolitan areas).
• A T1 Outstate System has been developed for
longer distances between cities.
• It's probable that your Internet access provider is
connected to the Internet as a point-of-presence
(POP) on a T1 line owned by a major telephone
network.
– POP: Locations where an Internet Service Provider
offers access to its network.
CSIT 220 (Blum)
36
CSU/DSU
• Channel Service Unit/Data Service Unit.
• Required for both ends of a T1 or T3 line.
• The CSU and DSU perform different tasks.
They are combined into one unit.
• The CSU is a device that performs
protective and diagnostic functions, like:
– Lightning protection
– “Ones density:” if more than 15 consecutive 0’s
are sent, raises alarm or “stuffs”
– Loopback (used for testing the lines)
CSIT 220 (Blum)
37
CSU/DSU (Cont.)
• The DSU is a device that connects a
terminal to a digital line.
• Converts data, which is digital, to a form
appropriate for the T1 line, which is also
digital but uses different encoding, voltage
values, etc.
– A kind of modem.
CSIT 220 (Blum)
38
DSU/CSU (Fig 12.2)
CSIT 220 (Blum)
39
DSU/CSU
• The DSU/CSU combination may also take
on some other duties
– Multiplexing (putting many signals on one line)
– Data Compression
CSIT 220 (Blum)
40
Comparison shopping
• There are various competing technologies
for establishing long distance connections
• There are two basic criteria for the
comparison
– Transmission rate (a.k.a. capacity)
– Cost
CSIT 220 (Blum)
41
Comparing Rates
CSIT 220 (Blum)
42
Comparing Rates (Cont.)
CSIT 220 (Blum)
43
Local loop
• Connection between the phone company’s Central
Office and an individual home or business.
• To connect computers from an individual’s home
or small office, one uses
–
–
–
–
Modem
ISDN
DSL
Cable modem
CSIT 220 (Blum)
44
ISDN
• Integrated Services Digital Network
• CCITT/ITU standards for digital transmission over
ordinary telephone copper wire
– ITU International Telecommunications Union
– CCITT, (now ITU-T) is an body for generating
telecommunications standards.
• A service offered by telephone companies to
connect computers over long distances (MAN or
WAN)
– Since it’s telephone companies, that 64 Kbps rate will be
seen here too.
CSIT 220 (Blum)
45
ISDN (Cont.)
dx
• One can send voice, video or data at speeds
better than modems (modem’s best: 56
Kbps).
– Hence I for “integrated.”
• Uses ISDN adapter in place of modem.
• Uses a mixture of packet switching and
circuit switching.
CSIT 220 (Blum)
46
ISDN Services
• There are two standard levels of ISDN
service: Basic rate (interface) (for home use)
and Primary rate (interface) (for business
use).
• Basic rate:
– Consists of three channels, two B (bearer)
channels and one D (delta) channel.
– The B channels operate at 64 Kbps, are full
duplex and carry voice or data.
CSIT 220 (Blum)
47
ISDN Services (Cont.)
– The D channel operates at 16 Kbps, is full
duplex and carries “signaling” (control)
information used to request services from the B
channels.
– One can use one B channel for voice and the
second for data or they can be combined
(“bonded”) to give the effect of one 128Kbps
channel.
CSIT 220 (Blum)
48
ISDN Services
• Primary rate:
– 23 64-Kbps B channels
• Primary rate in Europe has 30 B channels
– 1 64-Kbps D channel, although sometimes
addition D channels are installed.
CSIT 220 (Blum)
49
B-ISDN
• The original ISDN is a baseband technology
(one signal).
• Broadband ISDN, B-ISDN transmits many
signals simultaneously across the same
medium. It has a transmission rates of 1.5
Mbps.
– uses fiber optic cable
– not widely available
CSIT 220 (Blum)
50
ISDN (dying technology)
• When this standard was developed, good
modems were operating at 10Kbps (now
they’re up to 56 Kbps) and alternatives like
DSL and cable modems did not exist.
• Today it is not a much used technology.
CSIT 220 (Blum)
51
DSL
• Digital Subscriber Line
• DSL uses fancy modulation schemes to improve
the transmission rate possible on basic copper
wires (POTS, plain old telephone service).
• Sometimes called “last-mile technologies” since
they are only used to connect a telephone
(switching) station to a home and are not used
between stations.
– The distance to the office can play an important role
CSIT 220 (Blum)
52
DSL
• Recall that a voice signal uses
approximately the lowest 4 KHz of
bandwidth on what is typically twisted pair
wire.
• It is possible to send higher frequency
signals simultaneously, that is to use FDM
(frequency division multiplexing).
CSIT 220 (Blum)
53
Discrete Multi-tone (DMT)
modulation
• While the spectrum over 4 KHz is available for
additional signals, the system (POTS) was not
explicitly designed for its use.
• Consequently there may be variations in the
quality of these higher frequency signals.
• Thus the modulation scheme, Discrete Multi-tone
modulation (DMT), is “adaptive.” It searches for
usable parts of the spectrum and varies the rates
within channels depending on their signal-to-noise
ratios.
CSIT 220 (Blum)
54
Adaptive
• For high S/N (good channels), DSL chooses a
modulation scheme that encodes many bits per
baud.
– Recall how Shannon limits Nyquist.
• For low S/N (poor channels), DSL chooses a
modulation scheme that encodes fewer bits per
baud.
• DSL does not guarantee a specific rate, but does
as well as conditions allow.
CSIT 220 (Blum)
55
Upstream/Downstream
Upstream
Data from user
USER
Downstream
Data to user
CSIT 220 (Blum)
56
Asymmetric DSL (ADSL)
• Typical Internet user’s transmission traffic is
asymmetric, not the same in the upstream and
downstream directions.
• The upstream traffic tends to consist of “requests”
which are small packets.
• The downstream traffic tends to consist of large
files (often multimedia files).
• ADSL builds in this asymmetry.
CSIT 220 (Blum)
57
Asymmetry
• The FDM (frequency division multiplexing) makes
a number of channels available in a single wire.
• More of those channels are used for sending signals
downstream than for sending signals upstream.
• Not good for businesses offering web services
which require a higher upstream capacity.
– They need Symmetric DSL.
CSIT 220 (Blum)
58
Upstream/Downstream
Asymmetry
Upstream
Up to 640 Kbps
USER
Downstream
Up to 6.144 Mbps
CSIT 220 (Blum)
59
ADSL
• ADSL divides the frequency range into 286
separate channels.
• 255 transmit downstream, 31 transmit upstream and
2 for control information.
• The carriers for these channels are spaced at 4.1325
KHz intervals to keep signals from interfering.
• The signal may then be “inverse multiplexed,” that
is one signal sent over many wires.
– Recall multiplexing is several signals on one line, so
inverse multiplexing is one signal on many lines
(channels).
CSIT 220 (Blum)
60
Other DSL Technologies
• Symmetric DSL (SDSL)
– Symmetric transmission rates.
– Encodes differently so it sometimes works where ADSL
does not.
• High-Rate DSL (HDSL)
– 1.544 Mbps in both directions (same as T1).
– Short distances only.
– Requires 2 twisted pair wires, has some fault tolerance
since it can operate at reduced speeds with one wire.
CSIT 220 (Blum)
61
Other DSL Technologies
• Very-high bit rate DSL (VDSL)
– Up to 52 Mbps.
– Very restrictive on distances, so one comes into
neighborhoods (to an Optical Network Unit,
ONU) with fiber and then on to the nearby
houses with copper wiring.
CSIT 220 (Blum)
62
Other References
• http://www.webopedia.com
• http://www.whatis.com
• Understanding Data Communications &
Networks (William Shay)
• Signals, The Science of
Telecommunications (Pierce and Noll)
• Computer Dictionary, Mitchell Shnier
CSIT 220 (Blum)
63