How Power Spectrum Measurements Work

In a Vector Signal Analyzer (FFT Fast Fourier Transform: A mathematical operation performed on a time-domain signal to yield the individual spectral components that constitute the signal. See Spectrum. analyzer), the maximum span that you can accurately measure is limited by the sample rate. Nyquist theory dictates that the complex sample rate must be greater than the signal bandwidth. For example, in the M9393A Performance Vector Signal Analyzer, the maximum complex sample rate is 200 MHz Megahertz: A unit of frequency equal to one million hertz or cycles per second.. To account for filter-transition regions, the bandwidth is limited to 80 percent of the complex sample rate and therefore the usable frequency span is 160 MHz.

The Power Spectrum measurement extends the maximum span by making multiple FFT measurements at different LO frequencies. The Power Spectrum measurement steps the LO across the specified span, and at each step acquires a block of time samples that is then transformed into the frequency domain with an FFT operation. After the block of time data is acquired, the LO takes a step less than or equal to the bandwidth just processed, gathers another block of samples and computes the next FFT. This process repeats until it reaches the end of the measurement span. The multiple FFT measurements are then concatenated to achieve spans larger than that possible in typical FFT analyzers.

The size and resolution of each stepped FFT is a function of a few basics:

• sample rate
• number of samples in the time block and the associated resolution in the frequency domain
• shape and selectivity of the chosen window function
• desired frequency span
• specified RBW Resolution Band Width (RBW or ResBW): specifies the minimum frequency bandwith that two separate frequency spectra can be resolved and viewed seperately. For FFT (digital) based VSA's the process is equivalent to passing a time-domain signal through a bank of bandpass filters, whose center frequencies correspond to the frequencies of the FFT bins. For a traditional swept-tuned (non-digital) spectrum analyzer, the resolution bandwidth is the bandwidth of the IF filter which determines the selectivity.

The number of steps, or segments, used to create the power spectrum measurement depends on the frequency span and the resolution bandwidth (RBW). For higher ratios of RBW/span, the power spectrum measurement is able to compute the entire frequency spectrum without stepping its LO. For lower ratios of RBW/span, the measurement must step its LO two or more times to compute the entire frequency spectrum-in this case, the frequency spectrum is made up of 2 or more segments.

The Power Spectrum measurement is a scalar measurement--it does not include phase information. Because the measurement "pieces" together the frequency spectrum, it is seldom used for transient signals (transients that occur outside of the segment being evaluated are missed). Instead, power spectrum measurements are best for characterizing stationary signals. In fact, the measurement provides the same results as swept analyzers but with additional capabilities. For low-resolution bandwidths, power spectrum measurements are much faster than swept analyzers.

Use the Power Spectrum measurement when you need:

• Very narrow resolution bandwidth with high speed

• Very low noise floor with wide spans

• High signal-to-noise dynamic range

• Maximum flexibility of frequency span, resolution bandwidth, and speed

• Wider frequency spans

Specific examples of applications for which you would use scalar measurements include:

• Looking at spurious signals or noise

• Looking for low level distortion products