Signal-to-noise Ratio (SNR), Dynamic Range, and Noise

What are the differences between SNR (Signal to Noise Ratio), Dynamic Range, and Noise in the Presence of Signal?

These are all measurements that attempt to characterize the amplitude range of a device. That is the range from the lowest to the highest level signal a device can produce. These measurements are all typically expressed in dB.

First, for all of these measurements it is important to establish the full scale level of a device under test. For typical devices this is defined as the maximum level at which a device has 1% or less distortion. For these types of devices you adjust the level of the input signal or the gain of the device until you have determined this full scale reference signal. For many digital audio devices, this may simply be the maximum output level of the device. Regardless of the method used for determining the full scale reference level, Audio Precision instruments provide a convenient dBr function to store this level and allow all subsequent measurements, including noise measurements, to be reported relative to it.

In a conventional SNR measurement a device is first stimulated with a signal at full scale with its volume control set to maximum, if present. A level measurement is made to establish the maximum output reference. The stimulus tone is then removed and the inputs of the device are either terminated or shorted. The residual noise is then measured. The SNR value is then the ratio of the full scale output level of the device to the residual noise level of a device.

However, most digital audio devices mute their outputs when no signal is present. This would result in an inaccurately high SNR value. Instead the preferred measurement for characterizing digital to analog and analog to digital converters is dynamic range. To make a dynamic range measurement a device is stimulated with a full scale signal and then a tone that is 60 dB below the level of the full scale tone. The stimulus tone is then filtered from the output of the device with a sharp notch filter and the level of the remaining noise products is expressed as a ratio to the level of the full scale signal.

Signal to noise ratio and dynamic range are usually the same for linear devices such as amplifiers, but for time-varying devices such as systems using dynamic compression or noise reduction, or digital systems using floating point representation, dynamic range is typically greater than SNR.

Noise in the presence of signal is a modern test technique that is based on DSP/FFT analysis techniques to specifically analyze the output of a device and distinguish the original stimulus tone, its harmonic products, and the spurious noise products from each other. Noise in the presence of signal measurements are typically made by stimulating a device with a single or multi-tone signal and then analyzing the output of the device with a digital signal processing algorithm that produces a ratio of the energy in the noise of the output to the signal components of the output.

SNR essentially characterizes the ratio between the full scale output of a device and its idle noise. Dynamic range characterizes the ratio between the full scale output of a device and the spurious noise products created when a device is producing a very low level signal. Noise in the presence of signal characterizes the noise products that a device creates when it is reproducing a full scale signal; moreover it can be measured with a complex stimulus tone.

Audio Precision’s ATS-2 and 2700 series analyzers can make all of these measurements. SNR is available using a Quick launch macro, dynamic range measurements rely on the THD+N amplitude function meter, and noise in the presence of signal measurements can be made using our multi-tone analyzer.

As an additional note, our recent Technote #102, “Testing Personal Media players using QuickTest”, implements noise in the presence of signal measurements for MP3 players and other playback only devices.