Digital Audio: Part 3 - The Mathematics Of Sampling

Digital audio relies completely on sampling and no treatment of the subject can be complete without looking at how it works.

Fig.1. Sampling can be thought of as the audio waveform amplitude modulating a series of narrow pulses or needles, to produce a pulse-amplitude-modulated (PAM) signal.

Fig.2. Here, two different waveforms are represented by the same set of samples. Which is correct depends on the associated filters.

Fig.3. All sampling processes must be preceded and followed by appropriate filters. It is not meaningful to discuss sampling without considering the action of the filters.

Fig.4. When there is contra-rotation, as shown here, the horizontal processes are in opposition and cancel out, whereas the vertical process results in a sine wave having equal parts of positive and negative frequency. Modulation causes one sine wave to carry another so that negative frequencies subtract to produce lower side bands.

Fig.5. a) An un-modulated set of sampling needles has a spectrum like this. b) In conventional sampling, the base band is repeated as upper and lower sidebands around multiples of the sampling rate. c) The limit is reached at Fs/2, the Nyquist Frequency, when the lower sideband touches the baseband. d) An alternative sampling system in which the base band is band-limited. The lower side band is now below the baseband in the spectrum. The baseband can still be recovered with a band-pass filter.

Fig.6. In a reconstruction filter the zero crossings in the filter impulses align with adjacent samples so the output waveform effectively joins the samples into a continuous waveform identical within the available bandwidth to the original.

Fig.7. Transform duality holds that a square in either the frequency or time domains results in a sin/x in the other domain.

NAB 2021 Cancelled

NAB have announced the show scheduled for October 2021 has been cancelled.

PTP V2.1 – New Security & Monitoring For IP Broadcast Infrastructures - Part 1

Timing accuracy has been a fundamental component of broadcast infrastructures for as long as we’ve transmitted television pictures and sound. The time invariant nature of frame sampling still requires us to provide timing references with sub microsecond accuracy.

Creative Audio - Noise Reduction With Bob Bronow

Dialogue is king in television. Let’s face it, you don’t watch an episode of your favorite police procedural or reality show just to listen to the sound design or the incidental music. But whether the content is scripted or …

IP KVM - Delivering For Broadcasters - Part 2

In the last article in this series we looked at how KVM improves control, reliability, security and integration for multiple devices and cloud systems. In this article, we look at how latency is addressed so that users have the best…

Creative Audio - Broadcast Crowd Sound

When televised sports events began to return after the initial coronavirus lockdown in 2020, U.S. broadcasters faced a dilemma. With no spectators in attendance, what do you do about the lack of crowd noise? This is the fascinating story of…