The 9.6 dB comes from assuming a weakly nonlinear system to which Volterra series analysis can be applied. In reality, this simple approximation may not apply and higher order terms impact IIP3 and this gives a different number, often higher.
The old Dolby analog noise reduction scheme combined three of your favorite topics:
- envelope tracking
- compression (well, actually dynamic scaling)
- filtering
The idea was to use envelope tracking to allocate a portion for high frequencies (which tended to have lower power in the source as well as more hiss in the media), then use a variable amplifier (with one of those newfangled FET things acting as a variable resistor) to opportunistically increase the amount of gain and the filter turning point on the high pass.
The real trick was to correctly match the envelope tracking on the replay so the filter emphasis could be correctly reversed. That was never perfect, because the initial compression was based on the input signal and the recipient had to reconstruct the input only after knowing how to reverse the compression, but in practice Dolby found a scheme that converged well enough to satisfy listeners, especially as an alternative to the hiss of standard tape. Over time later schemes got better, and these days the licensing revenues built Dolby into a large business that is in video as well as audio.
This is why I love writing on the internet. Never in a million years would I have put those three things so nicely into a piece of history. Thank you for sharing!
Thanks for the great write-up.
I had a question around the 9.6 dB difference though.
Generally when we implement in circuit from simulation data we see that the difference between output P1dB and OIP3 is more than 10 dB.
Can you please throw some lights in that?
The 9.6 dB comes from assuming a weakly nonlinear system to which Volterra series analysis can be applied. In reality, this simple approximation may not apply and higher order terms impact IIP3 and this gives a different number, often higher.
Thanks for such an insightful write-up. It was quite clear, crisp, and concise.
Glad you liked it!
The old Dolby analog noise reduction scheme combined three of your favorite topics:
- envelope tracking
- compression (well, actually dynamic scaling)
- filtering
The idea was to use envelope tracking to allocate a portion for high frequencies (which tended to have lower power in the source as well as more hiss in the media), then use a variable amplifier (with one of those newfangled FET things acting as a variable resistor) to opportunistically increase the amount of gain and the filter turning point on the high pass.
The real trick was to correctly match the envelope tracking on the replay so the filter emphasis could be correctly reversed. That was never perfect, because the initial compression was based on the input signal and the recipient had to reconstruct the input only after knowing how to reverse the compression, but in practice Dolby found a scheme that converged well enough to satisfy listeners, especially as an alternative to the hiss of standard tape. Over time later schemes got better, and these days the licensing revenues built Dolby into a large business that is in video as well as audio.
This is why I love writing on the internet. Never in a million years would I have put those three things so nicely into a piece of history. Thank you for sharing!