Time Domain versus Frequency Domain
Figure 1: Three-dimensional illustration of the time and frequency domains
(Note: In the time domain, the three sinusoidal oscillations are shown separately, as on a three-channel oscilloscope.
A single-channel oscilloscope, however, displays only the instantaneous sum of their amplitudes when the three oscillations are superimposed.)
Figure 1: Three-dimensional illustration of the time and frequency domains
(Note: In the time domain, the three sinusoidal oscillations are shown separately, as on a three-channel oscilloscope.
A single-channel oscilloscope, however, displays only the instantaneous sum of their amplitudes when the three oscillations are superimposed.)
Time Domain versus Frequency Domain
The three-dimensional illustration in Figure 1 is used to compare the time domain and the frequency domain in signal processing. A signal consisting of several sinusoidal oscillations at different frequencies can be represented in both the time domain and the frequency domain. In the time domain, the variation of a signal over time is examined, as shown, for example, on an oscilloscope. In the frequency domain, by contrast, the frequency components contained in the signal are displayed, as shown, for example, on a spectrum analyzer.
In the illustration, three sinusoidal oscillations at different frequencies are shown on separate channels. When combined into a single channel, they are superimposed to form a composite signal. Its variation over time depends on the frequencies, amplitudes, and phases of the individual oscillations. With a suitable choice of these quantities, an approximately rectangular waveform can be produced, for example. The mathematical relationships are described by Fourier analysis.
The time domain (TD) describes how a signal quantity varies over time. The abscissa represents time. The ordinate usually represents a signal quantity such as electrical voltage. This representation is suitable, for example, for oscilloscopes and certain measurements made with modern network analyzers.
The frequency domain (FD) describes the frequency components of a signal. The abscissa represents frequency. The ordinate shows, for example, the amplitude or power of the corresponding frequency components. Such a representation can be obtained using spectrum analyzers, filter banks, or, in the simplest case, frequency-selective voltmeters.
The two viewing directions in Figure 1 are intended solely for illustration. In actual signal processing, a signal's frequency-domain representation is not obtained through geometric projection but through a mathematical transformation. A Fourier transform can be used to convert a signal from the time domain to the frequency domain. This determines both the amplitudes and the phases of the individual frequency components. Both types of information are required to reconstruct the original time-domain signal completely. The inverse transformation from the frequency domain back to the time domain is also possible.
The time domain and frequency domain are two equivalent, complementary representations of a signal. Which representation is more useful depends on the signal characteristics being investigated. A time-domain representation is particularly intuitive when examining waveforms, the timing of events, or propagation delays. In a pulse radar, the distance to a radar target can be determined directly from the time delay between the transmitted pulse and the echo signal. The classic display used for this purpose is the A-scope.
In modern radar systems, received signals are often additionally analyzed in the frequency domain. This makes it possible to determine Doppler frequencies, from which the radial velocities of targets can be inferred. Different Doppler frequencies also help distinguish moving targets from one another and from stationary objects. A more detailed analysis of the Doppler spectrum can reveal characteristic motion patterns that may be used, for example, for target classification. However, a target cannot be uniquely identified from a Doppler frequency alone.
In a linearly frequency-modulated FMCW radar, the distance to a target is determined from the frequency difference between the transmitted and received signals. This so-called beat frequency results from the propagation delay of the echo signal. For moving targets, it also contains a Doppler component. By appropriately processing several frequency ramps, range and radial velocity can be determined separately. Frequency-domain representation is therefore particularly important for signal processing in such radar systems.