Chapter 6:

Baseband Digital Transmission

1
Baseband Binary Pulse Amplitude Modulation Transmission System

Binary 𝑏 Pulse 𝑥 𝑡 Transmitting

data generator filter
source 𝑔 𝑡 𝐻 𝑓

Transmitter

Additive white
Gaussian noise
(AWGN)

𝑛 𝑡

Channel
𝐻 𝑓

Channel

Receiving 𝑟 𝑡 Synchronizer 𝑟 𝑡 Memoryless 𝑏 Binary

filter and decision data
𝐻 𝑓 sampler device sink

Receiver

2
 Nyquist Criterion

The condition for the removal of ISI is known as the Nyquist criterion. A necessary and sufficient condition for the
overall received pulse ℎ 𝑡 to satisfy

1, 𝑚 0
ℎ 𝑚𝑇
0, 𝑚 0

is that its Fourier transform 𝐻 𝑓 must satisfy

𝑘 1
𝐻 𝑓 𝑇 , 𝑓
𝑇 2𝑇

3
 Raised-Cosine Pulse (1)

- The Fourier transform of the pulse consists of a flat amplitude portion up to 1 𝛼 /2𝑇 and a cosine roll-off
portion between 1 𝛼 /2𝑇 and 1 𝛼 /2𝑇 , where 0 𝛼 1
‐ 𝐻 𝑓 is real and non-negative, that is, its phase spectrum is zero for all frequencies
- ℎ 𝑡 consists of the product of two terms, the first one is a sinc function through which zero crossings at all non-
zero multiples of 𝑇 are guaranteed, and the second one, which is a function of the roll-off factor 𝛼, significantly
reduces the tails and the predecessors of the pulse

1 𝛼
𝑇, 𝑓
2𝑇
𝜋𝛼𝑡
𝑡 cos
𝜋𝑇 1 𝛼 1 𝛼 1 𝛼 𝑇
𝐻 𝑓 𝑇 sin 𝑓 , 𝑓 & ℎ 𝑡 sinc
2𝛼 2𝑇 2𝑇 2𝑇 𝑇 4𝛼 𝑡
1
1 𝛼 𝑇
0, 𝑓
2𝑇

4
Raised-Cosine Pulse (2)

5
A Pulse Satisfying the Nyquist Criterion

6
 Example

Suppose the roll-off factor is 25% and the bandwidth of a baseband transmission system satisfying the Nyquist
criterion is 30 kHz. Determine the bit rate.

Solution

We have 𝑊 30,000 Hz, where 𝛼 25%. We thus obtain the bit rate 𝑟 48 kbps.

7
Required Bandwidth vs. Allowable Bit Rate vs. Potential Intersymbol Interference

1
bps
𝑇
𝛼 0 𝛼 1

𝐈 𝐈𝐈
2𝑥

Region I: Insufficient bandwidth  Unavoidable ISI

Region II: Practical bandwidth  Zero ISI
Region III: Wasted bandwidth  Zero ISI

𝐈𝐈𝐈

𝑊 Hz
0 𝑥

8
Eye Pattern (1)

9
 Eye Pattern (2)

Best sampling time

Slope: sensitivity
to timing error

a b

a: Time interval for sampling c: Noise margin at best sampling time

b: Distortion of zero crossings d: Distortion at sampling instant

10
 Eye Pattern (3)

Monitoring of an eye pattern can provide a qualitative measure of performance regarding the signal quality, including
the following important observations:

‾ The width of the eye opening represents the time interval during which the received signal can be sampled without
error from ISI
‾ The best time to sample the received signal is when the eye is open the widest
‾ The maximum distortion is indicated by the height of the eye opening at sampling time and it is twice the peak
distortion
‾ The noise margin or immunity to noise is defined by the height of the eye opening at the sampling time
‾ The sensitivity to timing errors is detected by the rate of closure of the eye as sampling time is varied
‾ Zero (level) crossings can provide clock information, and the amount of distortion of zero crossings indicates the
amount of jitter. The variation of level crossing can be seen from the width of the eye corners
‾ In a linear system with truly random data, all the eye openings would be identical
‾ Asymmetries in the eye opening indicate nonlinearities in the transmission channel
‾ When the effect of ISI is quite severe, traces from the upper portion of the eye pattern cross traces from the lower
portion, with the result that the eye is completely closed

11
 Binary vs M-ary (for info only)

Assuming both binary and 𝑀-ary signalling are delivering the same bit error rate, we can draw the following
conclusions:

i) The transmit power for the 𝑀-ary system must be increased by a factor of

ii) The bandwidth requirement for the 𝑀-ary system can be reduced by a factor of
iii) The complexity associated with the receiver for the 𝑀-ary system is higher than that for the binary system, as it
requires 𝑀 2 additional comparators.

In practice, if the bit rate must be higher than the Nyquist bandwidth, we then have no choice but to employ 𝑀-ary
signaling.

12
Bit Error Rate Performance for 𝑴-ary Signaling (for info only)

1.0E-1

𝑀 2
1.0E-2

𝑀 4
1.0E-3
𝑀 8

𝑀 16
1.0E-4
𝑒
𝑃

1.0E-5

1.0E-6

1.0E-7

1.0E-8
0 5 10 𝐸 15 20 25
dB
𝑁

13
 Linear Equalization

Input samples
𝑥 𝑘𝑇
𝑇 𝑇 𝑇

𝑐 𝑐 𝑐 𝑐

+ + + +
Output value
𝑟 𝑘𝑇
Error signal
𝑒 𝑘𝑇 𝑟 𝑘𝑇 𝑑 𝑘𝑇
Decision value
𝑑 𝑘𝑇

14
 Adaptive Equalization: LMS Algorithm (1)

- In order to determine the optimum tap values of an MMSE linear equalizer, a system of 2𝑁 1 linear
equations must be solved, but the inversion of a 2𝑁 1 2𝑁 1 matrix is computationally intensive
- Since the channel characteristics are always unknown and often time-varying, equalizers are designed to be
adaptive and the optimum tap values are thus obtained by an iterative procedure
- Adaptive equalization can track modest variations in the time-varying channels and continuously adjusts its
taps

15
 Adaptive Equalization: LMS Algorithm (2)

The Least Mean Square (LMS) algorithm is widely used, exhibits robust performance, and is as follows:
i. Set all tap values equal to zero
ii. Compute the output of the equalizer 𝑟 𝑘𝑇
iii. Compute the error signal 𝑒 𝑘𝑇 𝑟 𝑘𝑇 𝑑 𝑘𝑇 , the difference signal between the equalizer output 𝑟 𝑘𝑇
and the ideal (or estimated) output 𝑑 𝑘𝑇
iv. Compute the updated values of the taps, for a prescribed step size Δ 0 in accordance with the following relation:

𝑐 𝑘𝑇 𝑇 𝑐 𝑘𝑇 2Δ𝑒 𝑘𝑇 𝑥 𝑘𝑇

while noting 𝑥 𝑘𝑇 is the un-equalized received signal

v. Increment the time index 𝑘 by one, go back to step ii) and repeat the procedure until steady state conditions are
reached

16