* Voice over DSL: a great idea that just didn’t take
This time we’re continuing our discussion of packet voice technologies that seem to have fallen by the wayside. Last time we discussed voice over frame relay, so this time we’re moving on to voice over DSL.
Of course, we first have to say what we’re not talking about. Admittedly, there’s a lot of VoIP that goes over DSL – but in that case the IP and TCP/UDP layers are inserted in between the voice and DSL. Similarly, we’re not talking about the fact that residential analog voice tends to ride the same physical wire pairs that carry the DSL signal. Analog voice and DSL simply use different bandwidth, so they’re essentially using frequency division multiplexing.
Instead, VoDSL was proposed in recent years as a separate technology. The idea was quite simple and ingenious.
DSL technology typically includes using ATM as a “link layer” for framing. In fact, even residential ADSL services typically are deployed as “stealth ATM” with the TCP/IP traffic riding on top of ATM cells.
VoDSL was designed so that the terminating equipment would simply include a telephony interface to take analog voice and encode it as ATM cells that would be transported as a part of the “normal” DSL traffic. By using ATM AAL-1 for the transport, the voice traffic would be unquestionably “toll quality” and, further, would be totally compatible with the PSTN.
The technology was sound, and it worked well in trials. So what happened? First, just as with frame relay, there was no easy path to integrate this voice capability with IT infrastructure. Consequently, VoIP promises a much richer path for future evolution.
Additionally, we think there was a fundamental disconnect in the incumbent local exchange carriers. VoDSL was perfect for them. They are major players in the DSL market, and they specialize in local voice. However, for a plethora of reasons – from regulatory issues to protecting the installed base – this opportunity was not seized aggressively.
Consequently, VoDSL seems like yet another candidate to join SMDS in the list of great technologies that didn’t make it.
