23-05-2012, 12:11 PM
Digital Signal Processing
[attachment=22833]
Objectives:
This chapter introduces concepts of digital signal processing (DSP) and reviews
an overall picture of its applications. Illustrative application examples include
digital noise filtering, signal frequency analysis, speech and audio compression,
biomedical signal processing such as interference cancellation in electrocardiography,
compact-disc recording, and image enhancement.
1.1 Basic Concepts of Digital Signal Processing
Digital signal processing (DSP) technology and its advancements have dramatically
impacted our modern society everywhere. Without DSP, we would not
have digital/Internet audio or video; digital recording; CD, DVD, and MP3
players; digital cameras; digital and cellular telephones; digital satellite and TV;
or wire and wireless networks. Medical instruments would be less efficient or
unable to provide useful information for precise diagnoses if there were no
digital electrocardiography (ECG) analyzers or digital x-rays and medical
image systems. We would also live in many less efficient ways, since we would
not be equipped with voice recognition systems, speech synthesis systems, and
image and video editing systems. Without DSP, scientists, engineers, and technologists
would have no powerful tools to analyze and visualize data and
perform their design, and so on.
The concept of DSP is illustrated by the simplified block diagram in
Figure 1.1, which consists of an analog filter, an analog-to-digital conversion
(ADC) unit, a digital signal (DS) processor, a digital-to-analog conversion
(DAC) unit, and a reconstruction (anti-image) filter.
As shown in the diagram, the analog input signal, which is continuous in
time and amplitude, is generally encountered in our real life. Examples of such
analog signals include current, voltage, temperature, pressure, and light intensity.
Usually a transducer (sensor) is used to convert the nonelectrical signal to
the analog electrical signal (voltage). This analog signal is fed to an analog filter,
which is applied to limit the frequency range of analog signals prior to the
sampling process. The purpose of filtering is to significantly attenuate aliasing
distortion, which will be explained in the next chapter. The band-limited signal
at the output of the analog filter is then sampled and converted via the ADC
unit into the digital signal, which is discrete both in time and in amplitude. The
DS processor then accepts the digital signal and processes the digital data
according to DSP rules such as lowpass, highpass, and bandpass digital filtering,
or other algorithms for different applications. Notice that the DS processor
unit is a special type of digital computer and can be a general-purpose digital
computer, a microprocessor, or an advanced microcontroller; furthermore, DSP
rules can be implemented using software in general.
With the DS processor and corresponding software, a processed digital
output signal is generated. This signal behaves in a manner according to the
specific algorithm used. The next block in Figure 1.1, the DAC unit, converts
the processed digital signal to an analog output signal. As shown, the signal is
continuous in time and discrete in amplitude (usually a sample-and-hold signal,
to be discussed in Chapter 2). The final block in Figure 1.1 is designated as
a function to smooth the DAC output voltage levels back to the analog signal
via a reconstruction (anti-image) filter for real-world applications.
In general, the analog signal process does not require software, an algorithm,
ADC, and DAC. The processing relies wholly on electrical and electronic
devices such as resistors, capacitors, transistors, operational amplifiers, and
integrated circuits (ICs).
DSP systems, on the other hand, use software, digital processing, and algorithms;
thus they have a great deal of flexibility, less noise interference, and no
Analog
filter
ADC
DS
processor
DAC
Reconstruction
filter
Analog
input
Analog
output
Band-limited
signal
Digital
signal
Processed
digital signal
Output
signal
F IGURE 1.1 A digital signal processing scheme.
2 1 I N T R O D U C T I O N T O D I G I T A L S I G N A L P R O C E S S I N G
signal distortion in various applications. However, as shown in Figure 1.1, DSP
systems still require minimum analog processing such as the anti-aliasing and
reconstruction filters, which are musts for converting real-world information
into digital form and digital form back into real-world information.
Note that there are many real-world DSP applications that do not require
DAC, such as data acquisition and digital information display, speech recognition,
data encoding, and so on. Similarly, DSP applications that need no ADC
include CD players, text-to-speech synthesis, and digital tone generators, among
others. We will review some of them in the following sections.
1.2 Basic Digital Signal Processing Examples in Block Diagrams
We first look at digital noise filtering and signal frequency analysis, using block
diagrams.
1.2.1 Digital Filtering
Let us consider the situation shown in Figure 1.2, depicting a digitized noisy
signal obtained from digitizing analog voltages (sensor output) containing
a useful low-frequency signal and noise that occupies all of the frequency
range. After ADC, the digitized noisy signal x(n), where n is the sample number,
can be enhanced using digital filtering.
Since our useful signal contains the low-frequency component, the highfrequency
components above that of our useful signal are considered as noise,
which can be removed by using a digital lowpass filter. We set up the DSP block
in Figure 1.2 to operate as a simple digital lowpass filter. After processing the
digitized noisy signal x(n), the digital lowpass filter produces a clean digital
signal y(n). We can apply the cleaned signal y(n) to another DSP algorithm for a
different application or convert it to the analog signal via DAC and the reconstruction
filter.
The digitized noisy signal and clean digital signal, respectively, are plotted in
Figure 1.3, where the top plot shows the digitized noisy signal, while the bottom
plot demonstrates the clean digital signal obtained by applying the digital lowpass
filter. Typical applications of noise filtering include acquisition of clean
DSP
Digital filtering
x (n) y (n)
Digitized noisy input Clean digital signal
F IGURE 1.2 The simple digital filtering block.
1.2 Basic Digital Signal Processing Examples in Block Diagrams 3
digital audio and biomedical signals and enhancement of speech recording,
among others (Embree, 1995; Rabiner and Schafer, 1978; Webster, 1998).
1.2.2 Signal Frequency (Spectrum) Analysis
As shown in Figure 1.4, certain DSP applications often require that time domain
information and the frequency content of the signal be analyzed. Figure 1.5
shows a digitized audio signal and its calculated signal spectrum (frequency
content), defined as the signal amplitude versus its corresponding frequency for
the time being via a DSP algorithm, called fast Fourier transform (FFT), which
will be studied in Chapter 4. The plot in Figure 1.5 (a) is a time domain display
of the recorded audio signal with a frequency of 1,000 Hz sampled at 16,000
samples per second, while the frequency content display of plot (b) displays the
calculated signal spectrum versus frequencies, in which the peak amplitude is
clearly located at 1,000 Hz. Plot © shows a time domain display of an audio
signal consisting of one signal of 1,000 Hz and another of 3,000 Hz sampled at
16,000 samples per second. The frequency content display shown in Plot (d)
0 0.005 0.01 0.015 0.02 0.025 0.03
−2
−1
0
1
2
Noisy signal
Amplitude
0 0.005 0.01 0.015 0.02 0.025 0.03
−2
−1
0
1
2
Amplitude
Time (sec)
Time (sec)
Clean signal
FIGURE 1.3 (Top) Digitized noisy signal. (Bottom) Clean digital signal using the digital
lowpass filter.
4 1 I N T R O D U C T I O N T O D I G I T A L S I G N A L P R O C E S S I N G
gives two locations (1,000 Hz and 3,000 Hz) where the peak amplitudes reside,
hence the frequency content display presents clear frequency information of the
recorded audio signal.
As another practical example, we often perform spectral estimation of a
digitally recorded speech or audio (music) waveform using the FFT algorithm
in order to investigate spectral frequency details of speech information. Figure
1.6 shows a speech signal produced by a human in the time domain and
frequency content displays. The top plot shows the digital speech waveform
versus its digitized sample number, while the bottom plot shows the frequency
content information of speech for a range from 0 to 4,000 Hz. We can observe
that there are about ten spectral peaks, called speech formants, in the range
between 0 and 1,500 Hz. Those identified speech formants can be used for
Analog
filter
ADC
DSP
algorithms
x(n) Time domain display Analog
input
Frequency content display
F IGURE 1.4 Signal spectral analysis.
0 0.005 0.01
−5
0
5
A Time (sec)
C D
B
Signal amplitude
0 0.005 0.01
−10
−5
0
5
10
Time (sec)
Signal amplitude
0 2000 4000 6000 8000
0
2
4
6
Frequency (Hz)
Signal spectrum
0 2000 4000 6000 8000
0
2
4
6
Frequency (Hz)
Signal spectrum
1000 Hz
1000 Hz
3000 Hz
F IGURE 1.5 Audio signals and their spectrums.
1.2 Basic Digital Signal Processing Examples in Block Diagrams 5
applications such as speech modeling, speech coding, speech feature extraction
for speech synthesis and recognition, and so on (Deller et al., 1993).
1.3 Overview of Typical Digital Signal Processing in Real-World Applications
1.3.1 Digital Crossover Audio System
An audio system is required to operate in an entire audible range of frequencies,
which may be beyond the capability of any single speaker driver. Several
drivers, such as the speaker cones and horns, each covering a different frequency
range, are used to cover the full audio frequency range.
Figure 1.7 shows a typical two-band digital crossover system consisting of
two speaker drivers: a woofer and a tweeter. The woofer responds to low
frequencies, while the tweeter responds to high frequencies. The incoming digital
audio signal is split into two bands by using a digital lowpass filter and a digital
highpass filter in parallel. Then the separated audio signals are amplified.
Finally, they are sent to their corresponding speaker drivers. Although the
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
104
−2
−1
0
1
2
104 Speech data: We lost the golden chain.
Sample number
Speech amplitude
0 500 1000 1500 2000 2500 3000 3500 4000
0
100
200
300
400
Frequency (Hz)
Amplitude spectrum
F IGURE 1.6 Speech sample and speech spectrum.
6 1 I N T R O D U C T I O N T O D I G I T A L S I G N A L P R O C E S S I N G
traditional crossover systems are designed using the analog circuits, the digital
crossover system offers a cost-effective solution with programmable ability,
flexibility, and high quality. This topic is taken up in Chapter 7.
1.3.2 Interference Cancellation in Electrocardiography
In ECG recording, there often is unwanted 60-Hz interference in the recorded
data (Webster, 1998). The analysis shows that the interference comes from
the power line and includes magnetic induction, displacement currents in leads
or in the body of the patient, effects from equipment interconnections, and
other imperfections. Although using proper grounding or twisted pairs minimizes
such 60-Hz effects, another effective choice can be use of a digital notch
filter, which eliminates the 60-Hz interference while keeping all the other useful
information. Figure 1.8 illustrates a 60-Hz interference eliminator using a
digital notch filter. With such enhanced ECG recording, doctors in clinics
can give accurate diagnoses for patients. This technique can also be applied
to remove 60-Hz interferences in audio systems. This topic is explored in depth
in Chapter 8.
1.3.3 Speech Coding and Compression
One of the speech coding methods, called waveform coding, is depicted in
Figure 1.9(a), describing the encoding process, while Figure 1.9(b) shows the
decoding process. As shown in Figure 1.9(a), the analog signal is first filtered by
analog lowpass to remove high-frequency noise components and is then passed
through the ADC unit, where the digital values at sampling instants are captured
by the DS processor. Next, the captured data are compressed using data
compression rules to save the storage requirement. Finally, the compressed
digital information is sent to storage media. The compressed digital information
Digital
audio x(n)
Digital
highpass filter
Digital
lowpass filter
Gain
Gain Tweeter:
The crossover passes
high frequencies
Woofer:
The crossover passes
low frequencies
F IGURE 1.7 Two-band digital crossover.
1.3 Overview of Typical Digital Signal Processing in Real-World Applications 7
can also be transmitted efficiently, since compression reduces the original data
rate. Digital voice recorders, digital audio recorders, and MP3 players are
products that use compression techniques (Deller et al., 1993; Li and Drew,
2004; Pan, 1985).
To retrieve the information, the reverse process is applied. As shown in
Figure 1.9b, the DS processor decompresses the data from the storage media
and sends the recovered digital data to DAC. The analog output is acquired by
filtering the DAC output via the reconstruction filter.
ECG recorder with
the removed 60 Hz
interference
ECG
preamplifier
60 Hz
interference
Digital notch filter for
eliminating 60 Hz
ECG signal interference
with 60 Hz
inteference
F IGURE 1.8 Elimination of 60-Hz interference in electrocardiography (ECG).
Analog
filter
ADC
DSP
compressor
Analog
input Storage
media
F IGURE 1.9A Simplified data compressor.
DSP
decompressor
DAC
Reconstruction
filter
Analog
output
Storage
media
F IGURE 1.9B Simplified data expander (decompressor).
8 1 I N T R O D U C T I O N T O D I G I T A L S I G N A L P R O C E S S I N G
1.3.4 Compact-Disc Recording System
A compact-disc (CD) recording system is described in Figure 1.10a. The analog
audio signal is sensed from each microphone and then fed to the anti-aliasing
lowpass filter. Each filtered audio signal is sampled at the industry standard
rate of 44.1 kilo-samples per second, quantized, and coded to 16 bits for each
digital sample in each channel. The two channels are further multiplexed and
encoded, and extra bits are added to provide information such as playing time
and track number for the listener. The encoded data bits are modulated for
storage, and more synchronized bits are added for subsequent recovery of
sampling frequency. The modulated signal is then applied to control a laser
beam that illuminates the photosensitive layer of a rotating glass disc. When
the laser turns on and off, the digital information is etched onto the photosensitive
layer as a pattern of pits and lands in a spiral track. This master disc forms
the basis for mass production of the commercial CD from the thermoplastic
material.
During playback, as illustrated in Figure 1.10b, a laser optically scans
the tracks on a CD to produce a digital signal. The digital signal is then
Left mic
Right mic
Anti-aliasing
LP filter
Anti-aliasing
LP filter
16-bit
ADC
16-bit
ADC
Multiplex
Encoding
Modulation
Synchronization
Optics and
Recording
F IGURE 1.10A Simplified encoder of the CD recording system.
CD
Optical pickup
Demodulation
Error correction
4
Oversampling
14-bit
DAC
14-bit
DAC
Anti-image
LP filter
Anti-image
LP filter
Amplified
left speaker
Amplified
right speaker
F IGURE 1.10B Simplified decoder of the CD recording system.
1.3 Overview of Typical Digital Signal Processing in Real-World Applications 9
demodulated. The demodulated signal is further oversampled by a factor of
4 to acquire a sampling rate of 176.4 kHz for each channel and is then passed
to the 14-bit DAC unit. For the time being, we can consider the oversampling
process as interpolation, that is, adding three samples between
every two original samples in this case, as we shall see in Chapter 12. After
DAC, the analog signal is sent to the anti-image analog filter, which is a lowpass
filter to smooth the voltage steps from the DAC unit. The output from each
anti-image filter is fed to its amplifier and loudspeaker. The purpose of the
oversampling is to relieve the higher-filter-order requirement for the antiimage
lowpass filter, making the circuit design much easier and economical
(Ambardar, 1999).
Software audio players that play music from CDs, such as Windows Media
Player and RealPlayer, installed on computer systems, are examples of DSP
applications. The audio player has many advanced features, such as a graphical
equalizer, which allows users to change audio with sound effects such as boosting
low-frequency content or emphasizing high-frequency content to make
music sound more entertaining (Ambardar, 1999; Embree, 1995; Ifeachor and
Jervis, 2002).
1.3.5 Digital Photo Image Enhancement
We can look at another example of signal processing in two dimensions. Figure
1.11(a) shows a picture of an outdoor scene taken by a digital camera on a cloudy
day. Due to this weather condition, the image was improperly exposed in natural
light and came out dark. The image processing technique called histogram equalization
(Gonzalez and Wintz, 1987) can stretch the light intensity of an
Original image
A B
Enhanced image
F IGURE 1.11 Image enhancement.
10 1 I N T R O D U C T I O N T O D I G I T A L S I G N A L P R O C E S S I N G
image using the digital information (pixels) to increase image contrast so that
detailed information in the image can clearly be seen, as we can see in Figure
1.11(b). We will study this technique in Chapter 13.
[attachment=22833]
Objectives:
This chapter introduces concepts of digital signal processing (DSP) and reviews
an overall picture of its applications. Illustrative application examples include
digital noise filtering, signal frequency analysis, speech and audio compression,
biomedical signal processing such as interference cancellation in electrocardiography,
compact-disc recording, and image enhancement.
1.1 Basic Concepts of Digital Signal Processing
Digital signal processing (DSP) technology and its advancements have dramatically
impacted our modern society everywhere. Without DSP, we would not
have digital/Internet audio or video; digital recording; CD, DVD, and MP3
players; digital cameras; digital and cellular telephones; digital satellite and TV;
or wire and wireless networks. Medical instruments would be less efficient or
unable to provide useful information for precise diagnoses if there were no
digital electrocardiography (ECG) analyzers or digital x-rays and medical
image systems. We would also live in many less efficient ways, since we would
not be equipped with voice recognition systems, speech synthesis systems, and
image and video editing systems. Without DSP, scientists, engineers, and technologists
would have no powerful tools to analyze and visualize data and
perform their design, and so on.
The concept of DSP is illustrated by the simplified block diagram in
Figure 1.1, which consists of an analog filter, an analog-to-digital conversion
(ADC) unit, a digital signal (DS) processor, a digital-to-analog conversion
(DAC) unit, and a reconstruction (anti-image) filter.
As shown in the diagram, the analog input signal, which is continuous in
time and amplitude, is generally encountered in our real life. Examples of such
analog signals include current, voltage, temperature, pressure, and light intensity.
Usually a transducer (sensor) is used to convert the nonelectrical signal to
the analog electrical signal (voltage). This analog signal is fed to an analog filter,
which is applied to limit the frequency range of analog signals prior to the
sampling process. The purpose of filtering is to significantly attenuate aliasing
distortion, which will be explained in the next chapter. The band-limited signal
at the output of the analog filter is then sampled and converted via the ADC
unit into the digital signal, which is discrete both in time and in amplitude. The
DS processor then accepts the digital signal and processes the digital data
according to DSP rules such as lowpass, highpass, and bandpass digital filtering,
or other algorithms for different applications. Notice that the DS processor
unit is a special type of digital computer and can be a general-purpose digital
computer, a microprocessor, or an advanced microcontroller; furthermore, DSP
rules can be implemented using software in general.
With the DS processor and corresponding software, a processed digital
output signal is generated. This signal behaves in a manner according to the
specific algorithm used. The next block in Figure 1.1, the DAC unit, converts
the processed digital signal to an analog output signal. As shown, the signal is
continuous in time and discrete in amplitude (usually a sample-and-hold signal,
to be discussed in Chapter 2). The final block in Figure 1.1 is designated as
a function to smooth the DAC output voltage levels back to the analog signal
via a reconstruction (anti-image) filter for real-world applications.
In general, the analog signal process does not require software, an algorithm,
ADC, and DAC. The processing relies wholly on electrical and electronic
devices such as resistors, capacitors, transistors, operational amplifiers, and
integrated circuits (ICs).
DSP systems, on the other hand, use software, digital processing, and algorithms;
thus they have a great deal of flexibility, less noise interference, and no
Analog
filter
ADC
DS
processor
DAC
Reconstruction
filter
Analog
input
Analog
output
Band-limited
signal
Digital
signal
Processed
digital signal
Output
signal
F IGURE 1.1 A digital signal processing scheme.
2 1 I N T R O D U C T I O N T O D I G I T A L S I G N A L P R O C E S S I N G
signal distortion in various applications. However, as shown in Figure 1.1, DSP
systems still require minimum analog processing such as the anti-aliasing and
reconstruction filters, which are musts for converting real-world information
into digital form and digital form back into real-world information.
Note that there are many real-world DSP applications that do not require
DAC, such as data acquisition and digital information display, speech recognition,
data encoding, and so on. Similarly, DSP applications that need no ADC
include CD players, text-to-speech synthesis, and digital tone generators, among
others. We will review some of them in the following sections.
1.2 Basic Digital Signal Processing Examples in Block Diagrams
We first look at digital noise filtering and signal frequency analysis, using block
diagrams.
1.2.1 Digital Filtering
Let us consider the situation shown in Figure 1.2, depicting a digitized noisy
signal obtained from digitizing analog voltages (sensor output) containing
a useful low-frequency signal and noise that occupies all of the frequency
range. After ADC, the digitized noisy signal x(n), where n is the sample number,
can be enhanced using digital filtering.
Since our useful signal contains the low-frequency component, the highfrequency
components above that of our useful signal are considered as noise,
which can be removed by using a digital lowpass filter. We set up the DSP block
in Figure 1.2 to operate as a simple digital lowpass filter. After processing the
digitized noisy signal x(n), the digital lowpass filter produces a clean digital
signal y(n). We can apply the cleaned signal y(n) to another DSP algorithm for a
different application or convert it to the analog signal via DAC and the reconstruction
filter.
The digitized noisy signal and clean digital signal, respectively, are plotted in
Figure 1.3, where the top plot shows the digitized noisy signal, while the bottom
plot demonstrates the clean digital signal obtained by applying the digital lowpass
filter. Typical applications of noise filtering include acquisition of clean
DSP
Digital filtering
x (n) y (n)
Digitized noisy input Clean digital signal
F IGURE 1.2 The simple digital filtering block.
1.2 Basic Digital Signal Processing Examples in Block Diagrams 3
digital audio and biomedical signals and enhancement of speech recording,
among others (Embree, 1995; Rabiner and Schafer, 1978; Webster, 1998).
1.2.2 Signal Frequency (Spectrum) Analysis
As shown in Figure 1.4, certain DSP applications often require that time domain
information and the frequency content of the signal be analyzed. Figure 1.5
shows a digitized audio signal and its calculated signal spectrum (frequency
content), defined as the signal amplitude versus its corresponding frequency for
the time being via a DSP algorithm, called fast Fourier transform (FFT), which
will be studied in Chapter 4. The plot in Figure 1.5 (a) is a time domain display
of the recorded audio signal with a frequency of 1,000 Hz sampled at 16,000
samples per second, while the frequency content display of plot (b) displays the
calculated signal spectrum versus frequencies, in which the peak amplitude is
clearly located at 1,000 Hz. Plot © shows a time domain display of an audio
signal consisting of one signal of 1,000 Hz and another of 3,000 Hz sampled at
16,000 samples per second. The frequency content display shown in Plot (d)
0 0.005 0.01 0.015 0.02 0.025 0.03
−2
−1
0
1
2
Noisy signal
Amplitude
0 0.005 0.01 0.015 0.02 0.025 0.03
−2
−1
0
1
2
Amplitude
Time (sec)
Time (sec)
Clean signal
FIGURE 1.3 (Top) Digitized noisy signal. (Bottom) Clean digital signal using the digital
lowpass filter.
4 1 I N T R O D U C T I O N T O D I G I T A L S I G N A L P R O C E S S I N G
gives two locations (1,000 Hz and 3,000 Hz) where the peak amplitudes reside,
hence the frequency content display presents clear frequency information of the
recorded audio signal.
As another practical example, we often perform spectral estimation of a
digitally recorded speech or audio (music) waveform using the FFT algorithm
in order to investigate spectral frequency details of speech information. Figure
1.6 shows a speech signal produced by a human in the time domain and
frequency content displays. The top plot shows the digital speech waveform
versus its digitized sample number, while the bottom plot shows the frequency
content information of speech for a range from 0 to 4,000 Hz. We can observe
that there are about ten spectral peaks, called speech formants, in the range
between 0 and 1,500 Hz. Those identified speech formants can be used for
Analog
filter
ADC
DSP
algorithms
x(n) Time domain display Analog
input
Frequency content display
F IGURE 1.4 Signal spectral analysis.
0 0.005 0.01
−5
0
5
A Time (sec)
C D
B
Signal amplitude
0 0.005 0.01
−10
−5
0
5
10
Time (sec)
Signal amplitude
0 2000 4000 6000 8000
0
2
4
6
Frequency (Hz)
Signal spectrum
0 2000 4000 6000 8000
0
2
4
6
Frequency (Hz)
Signal spectrum
1000 Hz
1000 Hz
3000 Hz
F IGURE 1.5 Audio signals and their spectrums.
1.2 Basic Digital Signal Processing Examples in Block Diagrams 5
applications such as speech modeling, speech coding, speech feature extraction
for speech synthesis and recognition, and so on (Deller et al., 1993).
1.3 Overview of Typical Digital Signal Processing in Real-World Applications
1.3.1 Digital Crossover Audio System
An audio system is required to operate in an entire audible range of frequencies,
which may be beyond the capability of any single speaker driver. Several
drivers, such as the speaker cones and horns, each covering a different frequency
range, are used to cover the full audio frequency range.
Figure 1.7 shows a typical two-band digital crossover system consisting of
two speaker drivers: a woofer and a tweeter. The woofer responds to low
frequencies, while the tweeter responds to high frequencies. The incoming digital
audio signal is split into two bands by using a digital lowpass filter and a digital
highpass filter in parallel. Then the separated audio signals are amplified.
Finally, they are sent to their corresponding speaker drivers. Although the
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
104
−2
−1
0
1
2
104 Speech data: We lost the golden chain.
Sample number
Speech amplitude
0 500 1000 1500 2000 2500 3000 3500 4000
0
100
200
300
400
Frequency (Hz)
Amplitude spectrum
F IGURE 1.6 Speech sample and speech spectrum.
6 1 I N T R O D U C T I O N T O D I G I T A L S I G N A L P R O C E S S I N G
traditional crossover systems are designed using the analog circuits, the digital
crossover system offers a cost-effective solution with programmable ability,
flexibility, and high quality. This topic is taken up in Chapter 7.
1.3.2 Interference Cancellation in Electrocardiography
In ECG recording, there often is unwanted 60-Hz interference in the recorded
data (Webster, 1998). The analysis shows that the interference comes from
the power line and includes magnetic induction, displacement currents in leads
or in the body of the patient, effects from equipment interconnections, and
other imperfections. Although using proper grounding or twisted pairs minimizes
such 60-Hz effects, another effective choice can be use of a digital notch
filter, which eliminates the 60-Hz interference while keeping all the other useful
information. Figure 1.8 illustrates a 60-Hz interference eliminator using a
digital notch filter. With such enhanced ECG recording, doctors in clinics
can give accurate diagnoses for patients. This technique can also be applied
to remove 60-Hz interferences in audio systems. This topic is explored in depth
in Chapter 8.
1.3.3 Speech Coding and Compression
One of the speech coding methods, called waveform coding, is depicted in
Figure 1.9(a), describing the encoding process, while Figure 1.9(b) shows the
decoding process. As shown in Figure 1.9(a), the analog signal is first filtered by
analog lowpass to remove high-frequency noise components and is then passed
through the ADC unit, where the digital values at sampling instants are captured
by the DS processor. Next, the captured data are compressed using data
compression rules to save the storage requirement. Finally, the compressed
digital information is sent to storage media. The compressed digital information
Digital
audio x(n)
Digital
highpass filter
Digital
lowpass filter
Gain
Gain Tweeter:
The crossover passes
high frequencies
Woofer:
The crossover passes
low frequencies
F IGURE 1.7 Two-band digital crossover.
1.3 Overview of Typical Digital Signal Processing in Real-World Applications 7
can also be transmitted efficiently, since compression reduces the original data
rate. Digital voice recorders, digital audio recorders, and MP3 players are
products that use compression techniques (Deller et al., 1993; Li and Drew,
2004; Pan, 1985).
To retrieve the information, the reverse process is applied. As shown in
Figure 1.9b, the DS processor decompresses the data from the storage media
and sends the recovered digital data to DAC. The analog output is acquired by
filtering the DAC output via the reconstruction filter.
ECG recorder with
the removed 60 Hz
interference
ECG
preamplifier
60 Hz
interference
Digital notch filter for
eliminating 60 Hz
ECG signal interference
with 60 Hz
inteference
F IGURE 1.8 Elimination of 60-Hz interference in electrocardiography (ECG).
Analog
filter
ADC
DSP
compressor
Analog
input Storage
media
F IGURE 1.9A Simplified data compressor.
DSP
decompressor
DAC
Reconstruction
filter
Analog
output
Storage
media
F IGURE 1.9B Simplified data expander (decompressor).
8 1 I N T R O D U C T I O N T O D I G I T A L S I G N A L P R O C E S S I N G
1.3.4 Compact-Disc Recording System
A compact-disc (CD) recording system is described in Figure 1.10a. The analog
audio signal is sensed from each microphone and then fed to the anti-aliasing
lowpass filter. Each filtered audio signal is sampled at the industry standard
rate of 44.1 kilo-samples per second, quantized, and coded to 16 bits for each
digital sample in each channel. The two channels are further multiplexed and
encoded, and extra bits are added to provide information such as playing time
and track number for the listener. The encoded data bits are modulated for
storage, and more synchronized bits are added for subsequent recovery of
sampling frequency. The modulated signal is then applied to control a laser
beam that illuminates the photosensitive layer of a rotating glass disc. When
the laser turns on and off, the digital information is etched onto the photosensitive
layer as a pattern of pits and lands in a spiral track. This master disc forms
the basis for mass production of the commercial CD from the thermoplastic
material.
During playback, as illustrated in Figure 1.10b, a laser optically scans
the tracks on a CD to produce a digital signal. The digital signal is then
Left mic
Right mic
Anti-aliasing
LP filter
Anti-aliasing
LP filter
16-bit
ADC
16-bit
ADC
Multiplex
Encoding
Modulation
Synchronization
Optics and
Recording
F IGURE 1.10A Simplified encoder of the CD recording system.
CD
Optical pickup
Demodulation
Error correction
4
Oversampling
14-bit
DAC
14-bit
DAC
Anti-image
LP filter
Anti-image
LP filter
Amplified
left speaker
Amplified
right speaker
F IGURE 1.10B Simplified decoder of the CD recording system.
1.3 Overview of Typical Digital Signal Processing in Real-World Applications 9
demodulated. The demodulated signal is further oversampled by a factor of
4 to acquire a sampling rate of 176.4 kHz for each channel and is then passed
to the 14-bit DAC unit. For the time being, we can consider the oversampling
process as interpolation, that is, adding three samples between
every two original samples in this case, as we shall see in Chapter 12. After
DAC, the analog signal is sent to the anti-image analog filter, which is a lowpass
filter to smooth the voltage steps from the DAC unit. The output from each
anti-image filter is fed to its amplifier and loudspeaker. The purpose of the
oversampling is to relieve the higher-filter-order requirement for the antiimage
lowpass filter, making the circuit design much easier and economical
(Ambardar, 1999).
Software audio players that play music from CDs, such as Windows Media
Player and RealPlayer, installed on computer systems, are examples of DSP
applications. The audio player has many advanced features, such as a graphical
equalizer, which allows users to change audio with sound effects such as boosting
low-frequency content or emphasizing high-frequency content to make
music sound more entertaining (Ambardar, 1999; Embree, 1995; Ifeachor and
Jervis, 2002).
1.3.5 Digital Photo Image Enhancement
We can look at another example of signal processing in two dimensions. Figure
1.11(a) shows a picture of an outdoor scene taken by a digital camera on a cloudy
day. Due to this weather condition, the image was improperly exposed in natural
light and came out dark. The image processing technique called histogram equalization
(Gonzalez and Wintz, 1987) can stretch the light intensity of an
Original image
A B
Enhanced image
F IGURE 1.11 Image enhancement.
10 1 I N T R O D U C T I O N T O D I G I T A L S I G N A L P R O C E S S I N G
image using the digital information (pixels) to increase image contrast so that
detailed information in the image can clearly be seen, as we can see in Figure
1.11(b). We will study this technique in Chapter 13.