'
EE4601
Communication Systems
Lecture 2
Channel Capacity
&
c
2007,
Georgia Institute of Technology (lect2 0)
'
Shannon Capacity of a Channel
Claude Shannon (1949) proved that every physical channel has a capacity, C,
dened as the maximum possible rate that information can be transmitted with
arbitrary reliability over the channel.
Arbitrary reliability means that the bit error probability (BER) can be made
as small as desired.
Arbitrary reliability is achieved in practice by using forward error correction
(FEC) coding techniques.
Conversely, information cannot be transmitted reliably over a channel at any
rate greater than the channel capacity, C. The BER probability will be equal to
1/2 (useless channel).
The channel capacity depends on the channel transfer function and the received
bit energy-to-noise ratio (operating SNR).
&
c
2005,
Georgia Institute of Technology (lect2 1)
'
Coding Channel and Capacity
The channel capacity depends only on the coding channel, dened as the portion
of the communication system that is seen by the coding system.
The input to the coding channel is the output of the encoder.
The output of the coding channel is the input to the decoder.
The coding channel inputs are symbols chosen from a digital modulation alphabet, while the coding channel output are the matched filter outputs in the receiver.
Encoder
Coding
Channel
Decoder
&
c
2005,
Georgia Institute of Technology (lect2 2)
'
AWGN Channel
s(t)
Sn(f)
No /2
r(t) = s(t) + n(t)
-W
n(t)
For the AWGN channel (with continuous time input/output), the capacity is
C = W log2
P
1+
No W
W = channel bandwidth (Hz)
P = constrained input signal power (watts)
No = one-sided noise power spectral density (watts/Hz)
&
c
2005,
Georgia Institute of Technology (lect2 3)
'
Capacity of the AWGN Channel
Dividing both sides by W
P
C
= log2 1 +
W
No W
R
Eb
Eb /No
R/W
=
=
=
=
= log2
Eb R
1+
No W
1/T = data rate (bits/second)
energy per data bit (Joules) = P T
received bit energy-to-noise spectral density ratio (dimensionless)
bandwidth eciency (bits/s/Hz)
If R = C, i.e., we transmit at a rate equal to the channel capacity, then
Eb C
C
= log2 1 +
W
No W
or inverting this equation we get Eb /No in terms of C, viz.
&
2C/W 1
Eb
=
No
C/W
c
2005,
Georgia Institute of Technology (lect2 4)
'
AWGN Channel Capacity
&
c
2005,
Georgia Institute of Technology (lect2 5)
'
Capacity of the AWGN Channel
Example: Suppose that W = 6 MHz (TV channel bandwidth) and SNR =
P/(No W ) = 20 dB. What is the channel capacity?
Answer: C = 6 106 log2 (1 + 100) = 40 Mbps.
Asymptotic behavior: as C/W 0.
Using LHopitals rule
limC/W 0
Eb
=
No
=
=
=
limC/W 0 2C/W ln 2
ln 2
0.693
1.6dB
Conclusion: It is impossible to communicate on an AWGN channel with arbitrary
reliability if Eb /No < 1.6 dB, regardless of how small we make the bandwidth
eciency (or date rate).
&
c
2005,
Georgia Institute of Technology (lect2 6)
'
AWGN Channel Capacity
Power Ecient Region: R/W < 1 bits/s/Hz. In this region we have plenty
of bandwidth, but transmit power is limited, e.g., deep space communications.
Bandwidth Ecient Region: R/W > 1 bits/s/Hz. In this region we have
plenty of power, but bandwidth is limited, e.g., commercial wireless communications.
Observe that most uncoded modulation schemes operate about 10 dB from the
Shannon capacity limit for an error rate of 105 .
State-of-the-art turbo coding schemes can close this gap to less than 1 dB,
with the cost of additional receiver processing complexity and delay.
Generally, we can tradeo power, bandwidth, processing complexity, delay.
&
c
2005,
Georgia Institute of Technology (lect2 7)