Before the Dawn of IP Telephony - Part 27Noise and then more noise (summer 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
When entering 2002, the development of IP telephony devices and parts was set into motion. The author becomes involved in the development of an indoor device for telecommunication carriers and was busy with the selection of used parts.
The summer of 2002 was the period when development of indoor devices for telecommunication carriers was gaining momentum. A number of innovative advancements in technology were needed to shift indoor devices for telecommunication carriers to the "full dissemination" phase. And keeping cost low was especially difficult compared to conventional VoIP gateway.
There were some projects that focused too much on cost that attention to technologies that differentiated our products from the rest was lost as a result. We placed emphasis on using our technologies for voice signal processing and packet processing. After careful examination, we found the ideal architecture that would be most successful in applying our belief from the various options that were available.
Innovation for smaller and cheaper products
During this process, we were under pressure of having to choose a certain key device. One was an innovative device with extremely low power consumption in a small package. The other had normal power consumption but had solid performance history in the market that guaranteed reliable operation. The prices presented by both device vendors were almost the same. The former showed potential for future technical advancements and was anticipated as the core device of a new platform of CPE for telecommunication carriers.
However, merits are always balanced out by demerits and I had an instinct that using this device would be risky. Although the latter was the least innovative in terms of technical breakthroughs, it provided stable performance at a reasonable cost. On the other hand, the price seemed to push the limit since it required cooling measures equivalent to conventional products on the device side.
Devices delivered to telecommunication carriers require near-perfect stability as infrastructure devices-absolutely no experimenting. In all honesty, I was not sure whether to bet my chances on the future by taking technical risks or to play it safe for the time being.
Drawn by merits and demerits
Concerning the former device, we decided to analyze the technical background supporting its innovative properties, and weigh it against the trade-off risks. This particular part vendor was based in the U.S. but was extremely honest in answering the questions we had. Maybe it was the agency that provided such excellent correspondence. But we were able to receive answers concerning the low power consumption and high voltage, details that led to the realization of a compact package, its structure, and technical issues that they paid particular attention to. The demerits we indicated as the trade-off were answered in a sincere manner based on facts without denial. We were visited by the company staff in the U.S. on a number of occasions and they were more than happy to discuss any matter we had in mind. And through such exchange, I started to feel a connection with the engineers.
My first instinct of the device being risky gradually faded as I began to understand the technical background and their peculiar attention to the technology. As a result, we became true fans through complete understanding of this device along with the demerits. Sure there were demerits, but they had something we did not. And we were able to compensate for the demerits. I came to believe that something good could be made. Despite the risks, my mind was set on using this part.
Generally speaking, I feel that something that looks appealing at a glance can never be trusted. This is because there is always a downside to the apparent appeal. It's what makes things "balance." But there is also the appeal that is born after having an understanding of the entire picture, including the demerits.
The latter may seem to be the better choice in terms of performance and stability of product supply if the prices were the same. However, innovation is not born this way. The chances for groundbreaking advancement increase drastically when taking a path that seems risky. If you have the technology to cover the demerits, it is better to make a bold decision even if it requires you to stretch a bit. Being defensive by making halfway decisions will not lead to innovation. You will only get back what you put into it.
And innovation was indeed something we needed to push indoor devices for telecommunication carriers into full motion. Compact size and low cost were critical for widespread adoption and we felt that the trend would continue to accelerate.
The effects of reducing power consumption is immeasurable including the value of the device itself, reduction in cost of the entire device and smaller size. It may be presumptuous for me to say this, but we chose future potential over immediate stability. I believed that despite the difficulties for the time being, it would be well worth it in the long run.
Completion of prototype in June 2002 followed by a rough passage
In June 2002, the first prototype was finished. It was unbelievably small compared to other corporate devices on the market. But the initial report from the person in charge when turning on the power for the first time was, "There's noise."
I hate to be technical, but this part had extremely high impedance as a downside to its low power consumption. High impedance had a weakness of easily being affected by noise. The merits and demerits of this device boiled down to this fact. However, this noise was within the expected range and the use of this device was decided upon assuming that this problem could be solved on our own.
Photo 1: Testing environment used for sound quality assessment
But after some time, there was another report saying, "The first noise disappeared, but we have a different noise now." And such conditions continued. There were so many types of noise that we had a hard time trying to come up with names for expressing them.
As a characteristic of noise problems, noise is sometimes found hidden behind another noise. When you take measures for noise that was not heard until then, you find yourself hearing even bigger noise. What we were experiencing was this type of cycle. And when you have entered such a cycle, it's time to stop being easygoing about what is or isn't in the "expected range."
Despite some comments on whether or not we had made the right choice, we ultimately executed seven types of noise measures. Although unimaginable efforts went into these measures, we were finally able too see the goal of realizing the function by the end of June. And we had a solid feeling that the result of a new innovation as an indoor device for telecommunication carriers was near completion.
However, the demerits behind the merits decided to go on a rampage during the middle of a reliability test.
... To be continued
