Lecture 1: Introduction to Signal

Processing Basic structure of DSP processing

„ 1.0 Introduction
„ The aim of this lecture is to introduce some of the concepts of digital
signal processing.

„ Give some example of signal that can be processed.

„ Discuss some areas where DSP has found applications.

Digital signal processing and its Benefits Attraction of the DSP

„ DSP accuracy is limited to the number of bits used to represent the
„ The types of signal that has application with DSP processing signal.
include:-
„ Speech processing, including telephony, automatic speech „ The performance of the DSP processing does not change due to
external effects such as temperature.
recognition, and speech enhancement.
„ Biomedical applications, such as processing ECG signals to „ DSP solution to problems offers lower cost, and low power
determine heart disease, EEG signals to determine brain activity consumption.
or dysfunction.
„ Sound and music, such as MP3 encoding. „ DSP systems are based on programmed devices and consequently this
„ Image processing, such as processing information from CAT offers great flexibility without the need for modifying hardware.
scans to construct images.
„ Radar and sonar processing, such as identification of targets, „ Function that other wise can not be implemented in the analogue
their bearing, the range and speed by its Doppler frequency. domain can be implemented in the digital domain. For example, filter
that have a linear phase or complex adaptive filters.

The disadvantages of using DSP Application Areas

„ DSP has found application in a wide range of areas where information can be
„ The speed and cost of implementing DSP solutions to a problem can handled in digital form or controlled by a digital processor.
become expansive. This is particularly true in the case when processing
signal with large bandwidths. Image processing
„ pattern recognition where an object is recognized.
Image enhancement where the images can be consider to be a two dimensional
„ Limitations of DSP techniques are due to ADC/DAC converters. Also, „
image and the noise is filtered out.
the DSP processor devices are not fast enough to perform the
Images can be scanned into a PC and software is used to improve the
mathematical processing. „
appearance of the picture.

„ The design time involved in developing a DSP solution can be time Instrumentation
consuming. „ spectral analysis where a signal of sampled the DSP processor uses
mathematical algorithms to transform the signal from the time domain to the
frequency domain.
„ DSP represent signal with a limited number of bits. The limitation can
degrade the performance of the system may result. „ Position and rate control, is in the case of tracking a target
„ Kalman filters in which the state variables relate to the position, velocity and
acceleration.

1
Speech processing Telecommunications
„ Speech recognition is an application where DSP is widely used. „ The applications of DSP in the telecommunication include echo
„ Speech utterance is sampled and converted to feature vectors that cancellation.
represent the feature of speech.
„ The detection of a spoken word is achieved using likelihood measures The signal is reflected from the receiver back to the transmitter
of a sequence of word segments called phonemes. The probability for cause the one speaking to hear a delay version of his own
sequence of phonemes is model using Hidden Markov Models (HMM). voice.
DSP technique can be used to subtract the echoed signal.
Military applications
„ DSP processing has found many applications in the military including Biomedical processing
encryption and de-encryption providing secure communications. „ Some applications where DSP is used in the medical field
„ Radar and sonar processing such as non-cooperative target include tomographic reconstruction, patient monitoring,
recognition, target bearing information, target tracking, and missile analyzing and extracting information EEG signals when the
guidance.
patient undergo seizures, analyzing ECG signal for the diagnosis
of various heart problems.

Some Examples of Real-World

Applications using DSP Solution
Digital audio mixing „ The digital audio equalizer consists of a set of controllable
„ This allows the mixing and monitoring the characteristics of multiple digital filters that are used to manipulate parts of the signal
audio inputs spectrum to achieve the desired sound.

„ This is followed by a mixing matrix that allows the user to

combine the any of the equalized signals.

„ After mixing further DSP processing can be done using the

postprocessing, which may include reverberation an
equalization.

Speech Recognition „ The second task is concern with the identification of the speech
utterance.

„ The speech is tested against the predetermine templates and a word is

identified when a match is found.

„ Speech recognition is not an easy task to accomplish since

environmental factors such as background noise affect the
„ The concept of speech recognition can be divided into two tasks. performance.
„ The first task is the training phase where the user speaks a set of
words that are to be recognized. „ The change of the DSP processing is to provide robust processing
„ The computer than converts the spoken word to a set of parameter algorithms.
and a template of the word in stored.

2
 Biomedical DSP Application:
The Cochlea Implants
„ One area where DSP has found application in the biomedical field is with
cochlea implants.

SYLLABUS: EE4000 SIGNAL

PROCESSING 3
„ Lecture 1 Introduction: Applications of signal processing „ Software Tools
„ Lecture 2 Analogue to Digital Conversion „ Lecture 17 Software Tools Visual DSP++
Lecture 3 Digital-to-Analogue Conversion
„
„ Lecture 18 Assembler Language
„ Z-transform „ Lecture 19 Assembly Language
„ Lecture 4 Z transform definition „ Frequency Analysis of Signals
„ Lecture 5 Z-transform properties and applications „ Lecture 20 Discrete Fourier Transforms
„ Digital Filters „ Lecture 21 FFT Decimation-in-time
„ Lecture 6 Introduction to Digital Filters „ Lecture 22 FFT Decimation-in-frequency
„ Lecture 7 FIR Window Method, FIR Optimal Method.
„
„ Lecture 8 FIR Frequency Sampling Method
„ Lecture 9 IIR Pole-Zero Placement Method, IIR Impulse Invariant Method „ Time Analysis of Signals
„ Lecture 10 Matched Z-transform, Bilinear Z-transform „ Lecture 23 Correlation
„ Lecture 11 Filter Structure Realisations (Parallel, Cascade) „ Lecture 24 Convolution
„ Lecture 12 Lattice Implementation „ Multirate Filter
„ Lecture 25 Multirate filters: Decimation and Interpolation
„ Some Hardware Consideration to DSP
„ Lecture 13 Introduction to DSP Architectures
„ Lecture 26 Multirate filter structures design
„ Lecture 14 Core Architecture „ Exam Preparation Revision
„ Lecture 15 IO Processor
„ Lecture 16 ADSP 21262 evaluation board.