Unfortunately, we're all working here with the limited technical information we can scrape from various online sources. Companies don't always clearly divulge all the information on how their technologies exactly work. Most companies like to impress their customers and potential customers with all kinds of marketing jargon associated to a technology that most of the time is not exclusive to their products but that they would like to differentiate from their competitors. Bit depth expansion, e.g. from 16 bits to 24 bits, and upsampling or an increase of sample rates by inserting additional samples derived mathematically between two existing samples, a process also often referred to as interpolation, e.g. an increase from 44.1 kHz/sec to 96 kHz/sec or even to 192 kHz/sec, is one such example in the digital audio industry. Other companies employ similar upsampling technologies like Anthem or Cambridge Audio that uses Anagram Technologies' Q5 upsampling and some producers implement it within their DACs.
Keeping in mind the above paragraph, Denon's AL32 Processing webpage
, describes three functions associated with the operation of their AL32 technology.
The first, "Function 1. High-bit up-conversion (Adaptive Line Pattern Harmonized Algorithm)", is their bit depth expansion that increases the depth from 16 to 32 bits.
The second, "Function 2. Advanced ALPHA Processing", sounds to me like their take on upsampling. They describe it in these terms:
"The 44.1-kHz sampling signals of a CD are oversampled by a factor of 16 to produce a smoother waveform. At this time, simply performing linear interpolation and increasing the data will not produce a signal waveform that exists in the natural world.
A waveform close to that of the original signal is achieved by inferring data interpolated from a large volume of data that should be reproduced before and after the data read from the CD.
By analyzing large volumes of sampling before and after data read from the CD and inferring and interpolating the points that should exist, it is possible to produce a smooth signal that is closer to the original sound."
The last technique, "Function 3. Adaptive digital filter (Automatic Low-Pass filter Harmonic Adjustment)", is a digital filtering scheme to attenuate the occurrence of ringing.
Another PDF brochure for the Denon AVP-A1HDCI A-V Processor, the DVD-A1UDCI Player and the POA-A1HDCI Power Amplifier
that explains and showcases some key Denon technologies, including AL32, and that I referenced in a post I wrote back in January
, describes AL32 on page 7 in the following manner:
"Advanced AL32 Processing Multi Channel, to further improve proprietary
technology that optimizes HD audio performance, used for the first time in
In 1993, Denon unveiled “AL” Processing, a technology that expanded 16-bit digital
signals recorded on CD to 20 bits to produce a waveform that was close to that of
the original sound. This marked the beginning of “AL” Processing’s history of original
waveform reproduction technology.
“AL” Processing evolved further with the advent of various new digital disc media
such as DVD and Super Audio CD. This evolution progressed from AL24 Processing and
AL24 PLUS Processing to Advanced AL24 Processing that achieved sampling frequency
expansion, where data is interpolated along the time axis
in addition to bit expansion.
Now with the appearance of Blu-ray, Denon has developed Advanced AL32
Processing that further expands the number of bits from 24 to 32 to maximize the
exceptional audio performance of Blu-ray. With distortion-free sonic details, accurate
sound localization, and rich low range, Advanced AL32 Processing is able to reproduce
the original sound with greater fidelity."
Also look at the "Figure 1. Denon Advanced AL32 Processing" from my January post
(taken from the PDF brochure above), we can see in that diagram that the top box represents the "AL32 Bit Expansion" and it's followed by an "Upsampling" box. Although they don't specifically mention the sample rate increase in either of the two sources I've provided above, since they use 192kHz/32-bit DACs in the products discussed, it would make sense for the final sample rate to naturally be 192 kHz in order to match the DAC's capability.
So with the available information, this is the best explanation I can deduce although it may not be completely accurate.