Digital signal processing

Abstract

Signals are traditionally classified as being analog (continuous-time), discrete-time (sample-data), or digital. A continuous-time signal has infinite precision in both the time- and amplitude-domain. Discretetime signals have infinite amplitude precision, but are discretely resolved in time (sampled). Digital signals are of finite precision in both the time (sampled) and amplitude (quantized). Digital signals are either synthesized by a digital system (e.g., computer), or digitized by quantizing the sample values of an analog signal using an analog-to-digital converter (ADC). A digital-to-analog converter (DAC) maps a digital signal into an analog signal. Signal processing refers to the science of analyzing, synthesizing, and manipulating audio, acoustic, speech, video, image, geophysical, radar, radio signals, plus a host of other signals, using mathematical and experimental methods. Signal may be scalars, of one, two, or M dimensions, of finite or infinite duration. Digital signal processing (DSP) refers to the processing of digital or digitized signals using digital technologies. DSP system elements can be linear or nonlinear and perform in the time (e.g., filter) or transform domain (e.g., frequency). The processing agents range from mathematical and statistical abstractions or processes, to tangible software and hardware systems. DSP systems must often be designed to meet restrictive real-time speed, precision, dynamic range requirements in multisignal, multisystem environments. The design of a DSP solution, therefore, requires a concurrent knowledge of signal processing methods and technology.