Overview
ABSTRACT
Signal processing provides us with models and methods that allow us to analyze and interpret, in the broadest sense, the information contained in any form of observation. This is often called communication science. It is an important tool in many different technical areas, including signal analysis, systems and signal modelling, signal filtering, etc. The wide variety of transforms and methods it has to offer helps deepen our understanding of the phenomena we observe, by giving us access to different interpretations. Numerical techniques are now invaluable in the effort to model observed systems and to build prototypes of processing systems. The following provides the basic knowledge of digital signal processing through the study of deterministic signals and the commonly used transforms.
Read this article from a comprehensive knowledge base, updated and supplemented with articles reviewed by scientific committees.
Read the articleAUTHORS
-
Gérard BLANCHET : Director of Studies at Telecom-Paristech
-
Maurice CHARBIT : Professor at Telecom-Paristech
INTRODUCTION
The word signal refers to a series of observations indexed by an ordered sequence of real or integer values, usually time. When the index is in (R), we speak of a continuous-time or analog signal. If it is in (N) or (Z), we speak of a discrete-time or digital signal.
Digital signal processing, or DSP for short, is the processing of discrete-time signals. As in many other fields, the emergence of DSP is linked to the development of powerful computing resources, which have opened up its field of application to increasingly complex problems. Just look at "consumer" applications such as "high-definition television", "digital broadcasting", "mobile telephony", "multimedia applications" and so on. All these services make extensive use of digital signal processing, implementing algorithms that sometimes require considerable computing power. Here are just a few examples:
background noise suppression during a telephone transmission from the passenger compartment of a car;
cell phone signal processing ;
image and speech coding in a videophone system ;
speech recognition and synthesis ;
mechanical part fault detection and preventive maintenance;
sonar location of fish shoals;
radar evaluation of target position and speed;
seismic profile inversion for oil exploration ;
vibration analysis of an oil platform;
analysis of electroencephalographic signals to aid medical diagnosis;
reception of information from the GPS satellite navigation system;
electronic synthesis of musical sounds ;
processing acoustic signals from multiple microphones, etc.
Some of these issues have not yet been satisfactorily resolved, and are still under investigation. In addition to the complexity of the problems it tackles, TNS offers great flexibility in the development phase . Tests are perfectly reproducible. There is no risk of drift in component characteristics, as is the case with analog processing. Does this mean that analog processing is irretrievably doomed? In fact, even if their share is constantly...
Exclusive to subscribers. 97% yet to be discovered!
Already subscribed? Log in!
KEYWORDS
discrete time signal | discrete time Fourier transform | digital filters
Digital signal processing — Deterministic signals
Article included in this offer
"Electronics"
(
268 articles
)
Updated and enriched with articles validated by our scientific committees
A set of exclusive tools to complement the resources
Bibliography
Exclusive to subscribers. 97% yet to be discovered!
Already subscribed? Log in!