Common wisdom decrees that digital-oscilloscope users must always guard against aliasing by choosing a sampling rate that is several times the highest frequency component of the signal you are measuring. But, if you are analyzing narrowband intermediate-frequency (IF) signals in radios, pagers, and wireless communicators, you can make aliasing work for you instead of against you.

Fig 1 shows a receiver's 455-kHz IF signal. The digital oscilloscope samples the signals at 5M samples/sec into a record length of 1000 samples. The digital oscilloscope's FFT computed the spectrum. The display in Fig 1 shows only the 455-kHz carrier without a hit of modulation. The display shows no modulation because the point-to-point frequency resolution of the FFT equals the oscilloscope's sampling rate divided by its record length--5 kHz, in this case. This frequency resolution is much too wide to show any of the necessary detail.

Fig 2shows the same digital-oscilloscope setup with its sampling rate slowed by a factor of 200 to only 25k samples/sec. The FFT's resolution is now 25 Hz, unmasking a 440-Hz (middle A) frequency modulation of the IF's carrier. To achieve this resolution at the original sample rate would require a record length of 200,000 points--far beyond the capacity of any reasonably priced, high-bandwidth digital oscilloscope.

The 455-kHz carrier "mixes" with integer multiples of the sampling rate. In this case, the key multiple is 450k samples/sec. This multiple creates an alias at 5 kHz. In other words, the original 455-kHz carrier frequency aliases down to 5 kHz, and 25k samples/sec is adequate to recover the information content of the audio modulation of this aliased IF.