posted on Feb, 20 2009 @ 09:02 AM
Usually, a spectrum analyzer displays a power spectrum over a given frequency range, changing the display as the properties of the signal change.
There is a trade-off between how quickly the display can be updated and the frequency resolution, which is for example relevant for distinguishing
frequency components that are close together. With a digital spectrum analyzer, the frequency resolution is Δν = 1 / T, the inverse of the time T
over which the waveform is measured and Fourier transformed. With an analog spectrum analyzer, it is dependent on the bandwidth setting of the
bandpass filter. However, an analog spectrum analyzer will not produce meaningful results if the filter bandwidth (in Hz) is smaller than the square
root of the sweep speed (in Hz/s), which means that an analog spectrum analyzer can never beat a digital one in terms of frequency resolution for a
given acquisition time. Choosing a wider bandpass filter will improve the signal-to-noise ratio at the expense of a decreased frequency resolution.
With Fourier transform analysis in a digital spectrum analyzer, it is necessary to sample the input signal with a sampling frequency νs that is at
least twice the highest frequency that is present in the signal, due to the Nyquist limit. A Fourier transform will then produce a spectrum containing
all frequencies from zero to νs / 2. This can place considerable demands on the required analog-to-digital converter and processing power for the
Fourier transform. Often, one is only interested in a narrow frequency range, for example between 88 and 108 MHz, which would require at least a
sampling frequency of 216 MHz, not counting the low-pass anti-aliasing filter. In such cases, it can be more economic to first use a superheterodyne
receiver to transform the signal to a lower range, such as 8 to 28 MHz, and then sample the signal at 56 MHz. This is how an analog-digital-hybrid
spectrum analyzer works.
For use with very weak signals, a pre-amplifier can be used, although harmonic and intermodulation distortion may lead to the creation of new
frequency components that were not present in the original signal. A new method, without using a high local oscillator (LO) (that usualy produces a
high-frequency signal close to the signal) is used on the latest analyzer generation like Aaronia´s Spectran series. The advantage of this new method
is a very low noise floor near the physical thermal noise limit of -174 dBm.
(visit the link for the full news article)