Lecture 5 Information Theory
-
Upload
anthony-labriaga-canarias -
Category
Documents
-
view
221 -
download
0
Transcript of Lecture 5 Information Theory
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 1/51
Information Theory
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 2/51
- Bit rate is the number of bits per second.
- Baud rate is the number of signal units per second.
Baud rate is less than or equal to
the bit rate.
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 3/51
Example
An analog signal carries 4 bits in each
signal unit. If 1000 signal units are sent
per second.Find the baud rate and the bit rate.
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 4/51
Solution
Baud rate =Baud rate =
10001000 bauds per second (baud/s) bauds per second (baud/s) Bit rate = 1000 x 4 = 4000 bpsBit rate = 1000 x 4 = 4000 bps
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 5/51
Example
The bit rate of a signal is 3000. If
each signal unit carries 6 bits, what is
the baud rate?
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 6/51
Example
Baud rate = 3000 / 6 = 500 baud/sBaud rate = 3000 / 6 = 500 baud/s
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 7/51
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 8/51
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 9/51
Example
Find the minimum bandwidth for an ASK
system has a signal transmitting at 2000
bps. The transmission mode is half-duplex.
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 10/51
Solution
In ASK system the baud rate and bit rate
are the same. The baud rate is therefore
2000. An ASK signal requires a minimum bandwidth equal to its baud rate.
Therefore, the minimum bandwidth is
2000 Hz.
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 11/51
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 12/51
Solution
In ASK the baud rate is the same as the
bandwidth, which means the baud rate is
5000. But because the baud rate and the bit rate are also the same for ASK, the bit
rate is 5000 bps.
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 13/51
Quadrature amplitude modulation
( QAM) is a combination of ASK and
P SK so that a maximum contrast between each signal unit (bit, dibit,
tribit, and so on ) is achieved.
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 14/51
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 15/51
Bit and baud rate comparison
ModulationModulation UnitsUnits Bits/BaudBits/Baud Baud rateBaud rate Bit Rate
ASK, FSK, 2ASK, FSK, 2--PSK PSK Bit 1 N N
44--PSK, 4PSK, 4--QAMQAM Dibit 2 N 2N
88--PSK, 8PSK, 8--QAMQAM Tribit 3 N 3N
1616--QAMQAM Quadbit 4 N 4N
3232--QAMQAM Pentabit 5 N 5N
6464--QAMQAM Hexabit 6 N 6N
128128--QAMQAM Septabit 7 N 7N
256256--QAMQAM Octabit 8 N 8N
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 16/51
Information
What does a word information mean?
There is no some exact definition, however:
Information carries new specific knowledge,
which is definitely new for its recipient; Information is always carried by some specific
carrier in different forms (letters, digits, differentspecific symbols, sequences of digits, letters, and
symbols , etc.); Information is meaningful only if the recipient is
able to interpret it.
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 17/51
The information materialized is a message.
Information is always about something (size of aparameter, occurrence of an event, etc).
Viewed in this manner, information does not have
to be accurate; it may be a truth or a lie. Even a disruptive noise used to inhibit the flow of
communication and create misunderstandingwould in this view be a form of information.
However, generally speaking, if the amount of information in the received message increases,the message is more accurate.
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 18/51
Information
- More information means less
predications
- Less information means morepredications
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 19/51
Example
1. The sun will rise tomorrow
2. The FEU basketball team will win
against the Ateneo Basketball team
the next time they will play.
Question: Which sentence contains the most
information?
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 20/51
Information Theory
How we can measure the amount of
information?
How we can ensure the correctness of information?
What to do if information gets corrupted by
errors?
How much memory does it require to store
information?
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 21/51
Basic answers to these questions that formed
a solid background of the modern information
theory were given by the great American
mathematician, electrical engineer, and
computer scientist Claude E. Shannon in his
paper A Mathematical Theory of
Communication published in The BellSystem Technical Journal in October, 1948.
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 22/51
Claude Elwood Shannon (1916-2001)
The father of information theory
The father of practical digital circuit design theory
Bell Laboratories (1941-1972), MIT(1956-2001)
22
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 23/51
Information Content
What is the information content of any
message?
Shannons answer is: The information contentof a message consists simply of the number of
1s and 0s it takes to transmit it.
23
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 24/51
Information Content
Hence, the elementary unit of information is a
binary unit: a bit, which can be either 1 or 0;
true or false; yes or no, black and
white, etc.
One of the basic postulates of information
theory is that information can be treated like a
measurable physical quantity, such as densityor mass.
24
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 25/51
Information Content
Suppose you flip a coin one million times andwrite down the sequence of results. If you wantto communicate this sequence to another person,
how many bits will it take? If it's a fair coin, the two possible outcomes,
heads and tails, occur with equal probability.Therefore each flip requires 1 bit of information
to transmit. To send the entire sequence willrequire one million bits.
25
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 26/51
Information Content
Suppose the coin is biased so that headsoccur only 1/4 of the time, and tailsoccur 3/4. Then the entire sequence can
be sent in 811,300 bits, on average Thiswould seem to imply that each flip of the coin requires just 0.8113 bits totransmit.
How can you transmit a coin flip in lessthan one bit, when the only languageavailable is that of zeros and ones?
26
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 27/51
Obviously, you can't. But if the goal is totransmit an entire sequence of flips, and thedistribution is biased in some way, then you
can use your knowledge of the distribution toselect a more efficient code.
Another way to look at it is: a sequence of biased coin flips contains less "information"
than a sequence of unbiased flips, so it shouldtake fewer bits to transmit.
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 28/51
Information Content
Information Theory regards information as onlythose symbols that are uncertain to the receiver.
For years, people have sent telegraph messages,
leaving out non-essential words such as "a" and"the."
In the same vein, predictable symbols can be leftout, like in the sentence, "only information
essentially to understand must be tranmitted.Shannon made clear that uncertainty is the verycommodity of communication.
28
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 29/51
Information Content
Suppose we transmit a long sequence of onemillion bits corresponding to the first example.What should we do if some errors occur
during this transmission? If the length of the sequence to be
transmitted or stored is even larger that 1million bits, then 1 billion bits what shouldwe do?
29
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 30/51
Two main questions of
Information Theory
What to do if information gets corrupted by
errors?
How much memory does it require to storedata?
Both questions were asked and to a large
degree answered by Claude Shannon in his
1948 seminal article:
use error correction and data compression
30
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 31/51
Shannons basic principles of Information
Theory
Shannons theory told engineers how muchinformation could be transmitted over thechannels of an ideal system.
He also spelled out mathematically the principlesof data compression, which recognize what theend of this sentence demonstrates, that onlyinformation essentially to understand must betransmitted.
He also showed how we could transmitinformation over noisy channels at error rates wecould control.
31
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 32/51
32
Why is the Information Theory Important?
Thanks in large measure to Shannon's insights,
digital systems have come to dominate the world
of communications and information processing.
± Modems
± satellite communications
± Data storage
± Deep space communications
± Wireless technology
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 33/51
Information Theory
- Quantification of information
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 34/51
Channels
A channel is used to get informationacross:
Source
binary channel
0,1,1,0,0,1,1 Receiver
Many systems act like channels.
Some obvious ones: phone lines, Ethernet cables.
Less obvious ones: the air when speaking, TV screenwhen watching, paper when writing an article, etc.
All these are physical devices and hence prone to errors.
34
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 35/51
Noisy Channels
A noiseless binary channeltransmits bits without error:
0 0
1 1
A noisy, symmetric binary
channel applies a bit-flip
0m1 with probability p:
0 0
1 1
1p
1p
p
p
What to do if we have a noisy channel and
you want to send information across reliably?
35
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 36/51
Error Correction pre-Shannon
Primitive error correctionInstead of sending 0 and 1, send00 and 11.
The receiver takes the majority of the bit valuesas the intended value of the sender.
36
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 37/51
Channel Rate
When correcting errors, we have to be mindful of
the rate of the bits that you use to encode one bit
(in the previous example we had rate 1/3).
If we want to send data with arbitrarily small
errors,
then this requires arbitrarily low rates r, which iscostly.
37
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 38/51
Error Correction by Shannon
Shannons basic observations:
Correcting single bits is very wasteful andinefficient;
Instead we should correct blocks of bits.
38
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 39/51
A model for a Communication System
The communications systems are of a statistical
nature.
That is, the performance of the system can never
be described in a deterministic sense; it is always
given in statistical terms.
A source is a device that selects and transmits
sequences of symbols from a given alphabet. Each selection is made at random, although this
selection may be based on some statistical rule.
39
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 40/51
A model for a Communication System
The channel transmits the incoming symbols
to the receiver. The performance of the
channel is also based on laws of chance.
If the source transmits a symbol A, with a
probability of P{A} and the channel lets
through the letter A with a probability
denoted by P{A|A}, then the probability of transmitting A and receiving A is P{A}·P{A|A}
40
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 41/51
A model for a Communication System
The channel is generally lossy: a part of the
transmitted content does not reach its
destination or it reaches the destination in a
distorted form.
41
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 42/51
A model for a Communication System
A very important task is the minimization of
the loss and the optimum recovery of the
original content when it is corrupted by the
effect of noise.
A method that is used to improve the
efficiency of the channel is called encoding.
An encoded message is less sensitive to noise.
42
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 43/51
A model for a Communication System
Decoding is employed to transform the
encoded messages into the original form,
which is acceptable to the receiver.
Encoding: F : I F ( I )
Decoding: F -1: F ( I ) I
43
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 44/51
A Quantitative Measure of Information
Suppose we have to select some equipment
from a catalog which indicates n distinct
models:
The desired amount of information
associated with the selection of a particular
model must be a function of the
probability of choosing :
_ a1 2, ,..., n x x x
( )k I x
k x
k x
_ a ( )k k I x f P x!
44
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 45/51
A Quantitative Measure of Information
If, for simplicity, we assume that each one of
these models is selected with an equal
probability, then the desired amount of
information is only a function of n:
1 1/k I x f n!
45
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 46/51
A Quantitative Measure of Information
If each piece of equipment can be ordered in
one of m different colors and the selection of
colors is also equiprobable, then the amount
of information associated with the selection of
a color is :
_ a 2
1/ j j
I c f P c f m! !
jc
46
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 47/51
A Quantitative Measure of Information
The selection may be done in two ways:
Select the equipment and then the color
independently of each other
Select the equipment and its color at the same
time as one selection from mn possible
choices: & 1 /k j
I x c f mn!
1 1& ( ) ( ) 1 / 1 /k j k j I x c I x I c f n f m! !
47
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 48/51
A Quantitative Measure of Information
Since these amounts of information are identical,we obtain:
Among several solutions of this functionalequation, the most important for us is:
Thus, when a statistical experiment has neqiuprobable outcomes, the average amount of
information associated with an outcome is log n
log f x x!
1 / 1 / 1 / f n f m f mn !
48
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 49/51
A Quantitative Measure of Information
The logarithmic information measure has thedesirable property of additivity forindependent statistical experiments.
The simplest case to consider is a selectionbetween two eqiuprobable events. Theamount of information associated with theselection of one out of two equiprobable
events is and provides aunit of information known as a bit.
2 2log 1/ 2 log 2 1 ! !
49
8/3/2019 Lecture 5 Information Theory
http://slidepdf.com/reader/full/lecture-5-information-theory 50/51
Information Entropy
- A key measure of measure
- Expressed by the average number of
bits needed for storage orcommunication