In this month’s Whiteboard Session, Senior Product Management Specialist, Stuart Stephens explains wireless companding and how it works. Read the article below, or scroll down to watch the Whiteboard Video.
Companding is a process that happens in all analogue wireless systems to accommodate the limited dynamic range of FM radio.
The process takes its name from the compression and expansion process that occurs to achieve a full dynamic range signal on output. In other words, the signal is first compressed at the transmitter stage before it is expanded at the wireless receiver.
Lower-tier systems usually incorporate what’s known as a fixed ratio compander, while mid to high tier systems will utilise the more advanced form, known as audio reference companing. For the purpose of today’s topic, we’ll be covering the key differences between each of these companding types.
Fixed Ratio Companding
Through a fixed ratio compander, our audio signal is compressed at a fixed ratio (typically 2:1). This process is coupled with an expansion at the receiver with a ratio of 1:2, which restores our full dynamic range signal. The downside, of course, is that fixed ratio companding will perform the compression regardless of signal level. At low-level signals, the noise floor starts to become apparent, and this manifests itself in an audible artefact known as “breathing” or “pumping.”
Audio Reference Companding
Audio reference companding uses a soft-knee compression, which gradually introduces compression and allows the system to avoid the process until it’s absolutely necessary. Subsequently, these systems can avoid the low-level artefacts commonly associated with cheaper wireless systems. Additional benefits include lower system distortion and improved transients; overall, the system will sound more transparent and natural.
Digital Wireless Systems
Regular readers/viewers will remember from previous whiteboard sessions that digital wireless systems do not use or require any form of companding. In short, because we’re converting the audio signal into digital data at the transmitter there is no need to compress the signal — resulting in a potentially less coloured sound.