Before the Dawn of IP Telephony - Part 30IP is the key to high-quality telephones (fall 2002)
These contents translated a serialization article carried by ITPro IP telephony ONLINE published by Nikkei Business Publications, Inc. Jump to the original (Japanese).
Shinji Usuba
General Manager
eSound Venture Unit
Oki Electric Industry Co., Ltd
I was beginning to feel a change in quality of communication with the widespread deployment of broadband. And I had a hint for something new.
It was in fall of 2002 just when the development of the indoor device for telecommunication carriers was reaching completion. I was beginning to feel that there would be a new change in terms of communication quality when looking at the pattern of networks becoming full IP, and broadband lines spreading. This confirmed my belief for the latter of the two hypotheses mentioned previously: the hypothesis of improved communication quality.
When the ISDN network was being established, complete digitalization of the telephone network was the goal. In the case of ISDN, the infrastructure was structured by communication devices of various tiers. But the wall of "300 ~ 3.4kHz band" for voice existing from the early days could not be conquered. There was a concept of high-grade telephone even with ISDN and development was also made. In order to overcome the wall of the "300 ~ 3.4kHz band" restriction, however, all transmission networks had to be converted to broadband. This was not an easy wall to overcome to say the least.
In comparison, the infrastructure of an IP network is unbelievably simple, making it that much easier to overcome the wall needed to change voice quality. With networks moving into the direction of IP, I felt that now was the perfect time to improve voice quality using the characteristics of IP and changing the quality of communication.
We came to firmly believe that the wall of conventional telephone voice quality can be brought down by utilizing the potential of IP networks.
Hint from a broadband chip
Just when I was expanding such an idea, an opportunity to fully accelerate the verification of this hypothesis fell in our hands. And of all places, it came from the device vendor in the U.S. which we shared the pleasure and pain with during the development of the indoor device. I was given a single data sheet from a person in charge at the vendor's agency after a particular meeting.
"We have this chip..."
It was a data sheet introducing a subscriber circuit allowing a broad bandwidth to pass through.
Although I tried to not show it, something rang a bell inside my head. But the person in charge at the agency continued in a quiet, reserved tone, "Any chance you can make use of this for something?"
I asked, "Have you ever heard actual voice using this?"
"No," she responded, "we never had a telephone that allows you to hear voice in broadband..."
At that instant, the high-grade telephone developed during the ISDN period crossed my mind. And the established theory is that a technology missing the wave of the trend and branded "no good" don't usually come back to see the light of day.
But I had trouble swallowing this established theory.
To sidetrack a little, there was a time when examples from the past cellular phone field were dug up as materials for comparison during the examination of high-grade telephones in the IP period. Past marketing data indicates that promotion was made on the superiority of voice quality in the field of cellular phones, but did not bring forth results as expected. Marketing practices sometimes places emphasis on data. And I surely do not object to the use of past data of failure for analysis and to lead marketing in the right direction by not repeating the same mistake. However, it is important (and I have come to further believe this recently) that the person handling the data correctly understands the true reason why the product failed or did not become popular. Not correctly reading the data in many cases will lead to false conclusions. At times, there is a great chance that a technical product that failed in the past can be revived by changes in technical advancements and environment.
Let's bring the story back to high-grade telephones. Although high-grade telephones during the ISDN period didn't become widely accepted due to time and cost of building an infrastructure, the idea itself as a product is not bad. Even from the view standardizing voice encoding implemented at the time, dramatic improvement in voice quality must have been expected.
I had a sudden urge to hear the voice quality of ISDN high-grade telephone myself. I searched around the company, but with no luck. ISDN high-grade telephone was a product that was examined more than 10 years ago. And surely there were no engineers around that had been developing it at that time.
Obtaining a broadband telephone
For a long time, telephones were strictly designed for use within the bandwidth of "300 ~ 3.4kHz." In other words, bands less than 300Hz or more than 3.4kHz cannot pass through the communication network as sound and are cut-off as noise. And the better the cut-off accuracy, the higher the performance of the telephone. What I wanted was a telephone that did not limit the playback bandwidth.
Thinking there must be a broadband handset used during the development of the ISDN high-grade telephone, I went through tremendous effort before finally finding one. But that's another story to be told at another time.
The first broadband voice
Through much more effort than I had expected, I was finally able to prepare both the broadband subscriber circuit and handset. After a week, there was a call from the laboratory saying that the setup was ready. What does broadband voice sound like? I headed to the laboratory full of anticipation. In the room, I found a telephone and subscriber circuit casually connected to each other.
Okay, let's give it a try, I thought. But against all expectations, the result was a total disappointment.
"What's this? This is no better than a normal telephone."
... To be continued
