Oral-History:Keiji Tachikawa

About Keiji Tachikawa

Tachikawa has worked for NTT since 1962, working largely in digital microwave and satellite technology. He is proudest of his methodology of systems design for digital microwave technology, and for his development of three visions for NTT: the Integrated Services Digital Network vision (1979), the VI&P (multimedia) vision (1990), and the MAGIC (mobile multimedia) vision (1999). He discusses telecommunications developments in Japan since the 1950s (automated telephone switching, fiber optics, microwave technology, mobile communications, satellite technology); the introduction of competition into the Japanese telecommunications market with the privatization of NTT in 1985; the effects, advantages, and disadvantages of that deregulation; the role of prominent Japanese engineers and business leaders in the field, such as Shigeru Yonezawa and Tadahiro Sekimoto; the future of telecommunications (the continuing coexistence of wired and wireless networks); and the future of communications applications (internet, multimedia). He also discusses his involvement with the IEEE and the Japanese engineering society, IEICE, and their role in Japanese and international engineering. He lauds IEEE’s role in providing standards for the telecommunications field.

About the Interview

Keiji Tachikwawa: An Interview Conducted by David Hochfelder, IEEE History Center, 11 September 1999

Interview # 362 for the IEEE History Center, The Institute of Electrical and Electronics Engineers, Inc.

Copyright Statement

This manuscript is being made available for research purposes only. All literary rights in the manuscript, including the right to publish, are reserved to the IEEE History Center. No part of the manuscript may be quoted for publication without the written permission of the Director of IEEE History Center.

Request for permission to quote for publication should be addressed to the IEEE History Center Oral History Program, 39 Union Street, New Brunswick, NJ 08901-8538 USA. It should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.

It is recommended that this oral history be cited as follows:

Keiji Tachikawa, an oral history conducted in 1999 by David Hochfelder, IEEE History Center, New Brunswick, NJ, USA.

Interview

Interview: Dr. Keiji Tachikawa

Interviewer: David Hochfelder

Date: 11 September 1999

Place: Four Seasons Hotel, New York City

Japanese communications technology, 1950s-1990s

Hochfelder:

Dr. Tachikawa, would you give a summary of the advances in communications technology in Japan since the 1950s?

Tachikawa:

The basis for the development of telecommunications beginning in Japan in the 1950s was the recovery from the Second World War. The number of telephone subscribers was cut in half during the war from 1.1 million. The attempt to recover that loss was the major challenge for the telecommunications industry, and that’s why a lot of new technologies were implemented. The introduction and installation of automated switching systems was especially important. Up until that time most of the switches were Type A or Type H. A major percentage of the switches were replaced by the crossbar type.

Hochfelder:

This is in telephone switching?

Tachikawa:

Yes, telephone switching.

Hochfelder:

Would you briefly explain the different types of switches?

Tachikawa:

Type A and Type H were a Strowger type. I think the readers will definitely know what it is. The crossbar type is still a mechanical type. It basically works by crossing bars. It really closes the contacts. I think the automation of all the switches was completed in Japan in 1976. At the same time, the automation of long distance calling was also completed all over Japan. Through this process I believe the number of subscribers increased to over 30 million. I can give you the specific year this was accomplished later. At the same time tremendous progress was made in telecommunications technology in Japan. A full system of wire line with coaxial technology was achieved thus allowing multiplexing, and 10,000 circuits per line became possible.

Fiber optics was introduced in 1981, and currently most long distance telephone services use fiber optics. The construction, installation and implementation of fiber optics in telecommunications is happening in the 1990s. Currently the number of telephone subscribers in Japan is over 60 million. Another important technology in transmission systems is microwave technology. This technology was first introduced in the late '50s, and in the '70s it became digitalized. Currently most of this technology being introduced is digital microwave. In the 1980s additional mobile telecommunications were introduced, which has attracted a lot of attention and is growing very rapidly in the '90s. In terms of more detail regarding mobile telecommunication, the first cellular type of portable or automobile telephone was introduced in 1979, and Japan was one of the first countries to introduce this. I think it was a little bit earlier than the United States, where it was introduced about 1980-81.

Hochfelder:

Yes.

Tachikawa:

In 1987 the current form of portable or cellular telephone was introduced, and from 1993 onward the digital cellular phone has become popular. Around this time the number of subscribers increased dramatically, and now the number of cellular phone subscribers in Japan is over 60 million. In the past five years we have had an increase of more than 10 million subscribers each year. Another important technology that had a fairly early introduction in Japan is satellite communications. Unfortunately, the land area in Japan is rather limited, so that satellite communication is not very useful. It is used mainly for communication in case of disaster. Two things that Japanese technology and research and development have contributed are microwave satellite technology and fiber optics.

NTT career overview; ISDN

Hochfelder:

Would you give a brief summary of your career with NTT and talk a little about your involvement with these developments?

Tachikawa:

I joined NTT in 1962, and spent my first six years there as a radio engineer. I was involved in the development of digital microwave technology as well as satellite technology. I was awarded my doctorate degree later in 1981 as a part of the achievement of that research work. After that I was involved in the implementation of the telecommunications network for about six years. This was a time when the fixed telephone or wire line had a tremendous growth. At that time we were making an annual investment of over $10 billion.

I attended MIT Business School from 1977-78. After returning home from the U.S., I was involved in the creation of future visions for NTT on two occasions. The first time was in 1979 when NTT came up with the vision for the Integrated Services Digital Network (ISDN) in Japan. ISDN was part of the ITU vision, and Japan was one of the earlier countries to get involved in this process. At that time the plan was laid out for development of ISDN by the year 2000, and a tremendous effort was made for early introduction of ISDN. Actual introduction was made in 1984. I think that now the whole world is going in the direction of ISDN. Perhaps the highest penetration of ISDN now is in Germany, and I believe the second highest penetration is in Japan. About 10 million subscribers have ISDN in Japan now. Penetration is still low, however, when comparing those 10 million with the current 60 million line wire telephone subscribers.

Unfortunately, penetration and implementation of ISDN in the U.S. has been slowing down. ISDN used to supply a transmission line of 128 kilobits, which is worth two lines of telephone transmission into each household. However, with the advent of fiber optics in the 1980s there was another vision called Broadband ISDN (B-ISDN), which is an advanced formula ISDN. NTT also created a vision towards this end in 1990. This vision was called Visual, Intelligent and Personal (VI&P). That vision was introduced in the U.S. and picked up by Albert Gore, Vice President of the United States, and that has probably formed the basis of the current National Information Infrastructure (NII). Since he was appointed Vice President of the United States, he had proposed NII. Although the name of NTT’s vision was VI&P, in laymen’s terms in every language it is known as multimedia. In the 1990s much effort has been made to promote and develop multimedia. That was my second involvement with the creation of a long term vision.

I have become involved in a third vision since joining NTT DoCoMo, which is something that was worked on two years before its coming to completion in March of 1999. This is a vision for mobile communication towards 2010 which we call MAGIC. The letters stand for the five pillars or concepts of the vision. M stands for “Mobile Multimedia”; A stands for “Anytime, Anywhere, Anyone”; G stands for “Global Mobility Support”; I stands for “ Integrated Wireless Solution”; and C stands for “Customized Personal Service.” These are the things that will need to be provided in the development of mobile communications in the future. Being a part of the creation of these three visions are perhaps the most unique features of my personal career. A fourth vision is the introduction of competition in Japanese telecommunications. I got involved in that as well. The study of this possibility started in 1981, but competition was not introduced into the Japanese market until 1985. At the same time, the public corporation telecommunications entity was privatized. The splitting of AT&T in 1984 was a strong influence. I was also involved in the creation of a new legal structure and system for this new competitive environment, and I was involved in the reorganization of NTT after privatization. These are some of the more interesting experiences in my career.

Digital microwave technology

Hochfelder:

Dr. Tachikawa, what technical achievements in your career give you the most pride?

Tachikawa:

Perhaps the development of the methodology of designing systems in digital microwave technology. There was a methodology for designing analog microwave circuits by that time, but there was nothing for digital – not even internationally. This Japanese development was the first in the world, and that methodology is still used today. Basically this methodology designates or allocates a certain amount of noise – thermal, distortion and interference noise – to each part of the equipment according to the amount of noise allowable for each of these parts for specific devices, transmission routes and so on.

Hochfelder:

Would you explain more about the technical details?

Tachikawa:

By that time everybody was familiar with thermal noise, and there was a certain formula already established. As thermal noise increases the error increases, and a theoretical relationship between the thermal noise and the bit error rate had already been established. Distortion noise occurs when you transmit the digital signal because of the non-linear circuits. My method allowed engineers to figure out how much noise there is in a transmission route depending on the properties of frequency characteristics. Interference noise is due to the interference of the radio wave in space, in the air, so that the question is how much interference can be allowed. Radio waves are controlled by filters which restrict to a certain bandwidth, so we can design the filter to minimize the interference.

Corporate development visions

Hochfelder:

What accomplishments in business have given you the most pride?

Tachikawa:

I believe the major achievements were the services that we came up with through my involvement with the first three long term visions I mentioned. In 1979 it was the ISDN vision, and in 1990 it was multimedia vision. I think the flowers from both of those visions are just now blooming. This has expedited or created growth of business in Japanese telecommunications. Through the creation of this new vision that I just got involved in with NTT DoCoMo I would like to see promotion of growth in the business of mobile communications.

NTT privatization; corporate competition

Hochfelder:

Would you explain why the Japanese government decided to privatize NTT and introduce competition into the market?

Tachikawa:

I believe that the liberalization that happened in the U.S. and U.K. was a strong impetus for the introduction of a policy of competition in Japan. I think this was also a result of tremendous advances in telecommunications technology. The construction of a telecommunications infrastructure no longer required a huge organization like national government. It was seen that it could be done by private corporations. The concept that by introducing competition one could really expedite and promote the progress of technology was another motivation. The deregulation of airlines is a great example. In the 1970s and '80s, there was severe competition in that industry. Yet another reason for introducing competition was that this would spur more diversified development, and indeed a wide variety of services were developed. At the outset the government was not very enthusiastic, but looking at the situations in the U.S. and U.K., the decision was made. Competition was introduced in Japan in 1985. The same kind of competition had been introduced in the U.K. in 1982, and then there was the breakup of AT&T in 1984 in the U.S. Japan did not lag very far behind those two countries in this regard.

Hochfelder:

What does the telecommunications market in Japan look like today, fifteen years after privatization?

Tachikawa:

After the introduction of competition, progress was seen especially in long distance and international telephone services, and then also in mobile telecommunications. Local telecommunications services did not enjoy such positive results, however. As in the U.S., local telephone network competition did not work very well, and it is still a near monopoly situation today. However there are recently developments coming in, such as the introduction of CATV and the radio type of local telephone service. These new companies would like to provide easy Internet access through CATV and wireless access. After fifteen years the share of the former national operator at the local level is still 99 percent, but with the long distance it is probably down to 60 percent. For international telecommunications, it’s down to 40 percent. Mobile communication is now at about 57 percent, in what once was basically a monopolized situation. Competition has not been succeeding in all fields of telecommunications.

Recently it has become apparent that the current demarcation is basically meaningless, and reorganization of the entire industry is beginning. Competition has progressed in Japan in certain fields, including local, long distance, international, mobile and pager telecommunication services. It was so narrowly divided that while competition progressed in each of these fields from the user’s point of view, this kind of division and demarcation really became rather ineffective and useless on the business level.

Hochfelder:

What in your opinion were the advantages and disadvantages of privatization and the introduction of competition?

Tachikawa:

Competition brought more advancement in technology and lowered the tariff. Those were big advantages. Also, many new services were introduced. One of the drawbacks, however, was that because it was so demarcated in a small field, there was a certain imbalance in tariffs between various fields. For example the pager business, once very popular, is now on the decline. After competition was introduced, one result was that many companies specializing in pager services alone were created. Many of these companies face the prospect of bankruptcy today. Before the introduction of competition in the U.S. and U.K., there was a process of balancing the various tariffs among the various fields, such as local, long distance and international telecommunications services. The tariffs were revised so that they would be on a cost basis. This was not done in Japan. The competition was introduced in Japan while the tariffs had a distorted structure. All around the world in telecommunications basically the long distance services made the most money, and this was cross-subsidizing the local telephone services. It is more or less common sense that before introducing the competition the local telephone charges should be raised and the long distance charges lowered. However, they didn’t do that in Japan. Consequently, nobody went into the business in local telephone services, while many went into the long distance telephone business where money could be made fairly easily. There was a perception that it would be easy to make money in telecommunications. Currently there are complaints that there is no real competition in local telephone services, and this is due to a square peg being placed in a round hole in the beginning. The effects of this are still being felt now. Rather than these being the disadvantages of competition, they are really the disadvantages experienced from way the competition was introduced.

Hochfelder:

Right.

Tachikawa:

Perhaps the real disadvantage of competition is when the industry becomes overly competitive. One example is that currently in Japan there are four licensed mobile telecommunications companies and competition has become severe to the point of being excessive. In mobile communications, there are three companies in PHS. PHS is somewhat similar to PCS in the United States. All in all there are seven carriers of mobile communications that are in competition in one region. Severe competition causes the lowering of prices and the lowering of tariffs also continues, so that business operations are having a difficult time meeting costs.

IEEE and Communications Society; Japanese engineering societies

Hochfelder:

Would you talk about your involvement with the various professional societies, including the IEEE and the IEEE Communications Society? I would also like to hear about the Japanese engineering societies in which you have been active.

Tachikawa:

I have been a member of the IEEE since 1965, and became a senior member in 1987. When I was involved in research and development, I participated in IEEE activities and utilized IEEE resources to get information. From 1975 to ’77 I was in New York City, and during that time I was actively participated in various IEEE meetings and various academic conferences. At that time Japan was still basically catching up with the rest of the world, and close observation of IEEE activities was very important for Japan. In order to have a counterpart of the IEEE, The Institute of Electronics, Information and Communication Engineers (IEICE) was created in Japan, and this group has more than 40,000 members today. I served as an officer in various capacities for this Japanese institute. In addition, there is a Tokyo Section of IEEE, and I believe that chapter has about 5,000 members. I was involved in activities with both of these groups. Japan as a whole has really tried to make a lot of effort and dedicated itself to developing telecommunications technology, and the IEEE has been a great venue for various activities for Japanese engineers. In the beginning, the findings or theses of Japanese engineers were first submitted to IEICE, and the best ones were submitted to IEEE. This was a tremendous opportunity for Japanese engineers. In the 1990s it has became much easier to travel, so that now even young researchers can attend IEEE meetings all around the world. Japanese engineers are no longer obliged to first submit findings to the Japanese Institute. In other words, they now can go directly to the IEEE. I believe that IEEE activities have been very effective. The membership of IEEE is spreading all over the world, so that really there already is a wide exchange of information and people networking through IEEE, and there is more to come. What is the membership now?

Hochfelder:

I think the IEEE has about 200,000 members worldwide. The IEEE tries to be an international organization, even though it is predominantly U.S. membership.

Tachikawa:

There are about 50,000 members in the Communications Society alone.

Hochfelder:

That’s correct.

Tachikawa:

In Japan there is tremendous interest and benefits for engineers in the IEICE and there are 40,000 members. There have been recent instances of the IEEE and the Japanese Institute jointly sponsoring conferences and meetings. The VT (Vehicular Technology) Conference was held in Tokyo in the year 2000. In 1987 the Globecom Conference was held in Tokyo, and in 1996 the NOMS (Network Operations and Management Symposium) Conference. I had the honor of being the keynote speaker when the IEEE sponsored the Globecom Conference in Singapore in 1995. When conferences like Globecom are held outside the United States, they try to call in a prominent figure in the region, and that works well as a stimulus for development in the regional industry and local telecommunications.

In the world of telecommunications, standardization has become very important, and so the contribution of IEEE in this regard has also been important. The role of IEEE in standardization is important because when these standardization activities are carried out in international bodies like ITU it takes so much time that often the outcome is not really appropriate for the current situation.

Another factor we cannot really ignore is the political influence in the activities of international bodies, whereas something like the IEEE might not be so burdened. We are hoping that the IEEE will come to have more influence in terms of standardization activities.

Japanese communications leaders

Hochfelder:

Would you talk a little bit about some of the other prominent Japanese engineers and business leaders in the field of communications and perhaps other members of NTT with whom you have worked?

Tachikawa:

NTT has active members in the IEEE today, and there are a lot of other people in the electronics industry in Japan who are quite active in IEEE activities. Two years ago the president of NTT was awarded a medal from IEEE.

Hochfelder:

That’s right.

Tachikawa:

And Dr. Shigeru Yonezawa, the former president of NTT, and the former chairman of NEC, Dr. Tadahiro Sekimoto, also received medals. A number of prominent Japanese scholars have been active in the IEEE. Many CEOs and presidents of Japanese companies in the electronics industry have been members of the IEEE. In the past, the presidents of all of the companies in the electronics industry had engineering backgrounds. Reorganization in Japanese industries has replaced many presidents and CEOs who were engineers in favor of other candidates with liberal arts and humanities backgrounds. NEC and Sony replaced engineering presidents in favor of presidents with business backgrounds. In that context, engineers have new challenges to surmount as policies change. I believe that engineers tend to be considered inadequate to the task of dealing with labor issues and so on.

Predictions for wireless and wired communications

Hochfelder:

You talked about your involvement with the vision for the year 2010. In the United States, many people have predicted that wired communications will disappear and that the world will become wireless. Do you agree?

Tachikawa:

That is not really possible. I believe that wire line and wireless will coexist. First of all, the wireless the spectrum is finite, so we really cannot cover everything with wireless. The fixed line with fiber optics has a tremendous advantage in terms of high-speed data transmission. Fiber optics does not have the disadvantage of being finite. In terms of transmission speed, the fixed line is always several times better than the wireless. Specifically, a major part of the fixed line now is ISDN, which is the 64 kbps (kilobits per second) × 2, so it’s 128 kbps. Wireless is still only 11.2 kbps. Recently we have been able to get a higher speed transmission over wireless, but still it is just 64 kbps. Currently we are involved in the standardization for next generation technology, IMT-2000 (International Mobile Telecommunications-2000 ), and the highest rate of transmission for wireless there will be 2 Mbps (Megabits per second ). However, in the high-speed vehicular environment maximum speed will be 384 kbps. Two megabits is achieved only in a stationary position. With the wireline, currently in Japan the rate is 1.5 Mbps. In the U.S. it is 1.5 Mbps, and in Europe it is 2 Mbps. Fiber optics can easily transmit 150 Mbps. Even with tremendous effort, wireless is still limited to only 2 Mbps, so fiber optics has 75 times the capacity of wireless. Therefore the wire line cannot be replaced by wireless. I think users will utilize these two modalities for different purposes. For things likehigh-speed transmission and high vision they will probably use the wire line, whereas with voice only transmission wireless might be more popular.

This is not the end of the road. In a fourth generation system they are talking about 20 Mbps transmission speed for the wireless. The target for implementation of this is around the year 2010. In other words, in the year 2001 when we introduce the system it will be 2 Mbps, and around the year 2020 it will reach 20 Mbps. I believe that there is a tremendous changeover in technology every ten years in mobile communication. In the other words, the first generation, 1979-80, was analog; 1992-93 was digital; 2001 will be the third generation; and 2010 will be the fourth generation. In that sense the year 2020 will be the fifth generation. We will have to start studying the fifth generation soon. For the fixed side we need substitute the cheaper substitute of fiber optics like xDSL. That would be the fastest one for awhile, and it will not really be necessary to think about the higher speed but rather attention needs to be focused on applications.

Fiber optics has not penetrated further in the U.S. because so much metallic cable with a certain capacity was already installed underground. Digging up those cables and replacing them with fiber optics is very costly. This is why there is a view that we should make more use of the metallic cables and try to find a larger capacity for transmission in what is already there and things like xDSL have been developed. In that sense Japan was fortunate, because rather than having wire installed underground we had it in the form of aerial cable. This has made replacing the old cable with optical fiber a bit easier. This has also been better for Japan aesthetically.

The Internet

Hochfelder:

You mentioned the possibility of new applications. Would you speculate on what sort of new communications applications might be available to consumers in the next decade?

Tachikawa:

We’ve always been involved in the development of applications, but the introduction of the Internet in 1993 was a major event in terms of application. Applications that can utilize the high-speed capacity of the Internet and combine visual images and text will be in great demand in the future. From earlier times there has been discussion of the transmission of moving pictures, for example using fiber optics in telephone communications. A simple visual type of communication device such as a video phone has not really caught on in popularity, so that it’s only being used in video conferencing facilities for the most part. There is a need for applications to cover more than just moving pictures. It’s very important to use the visual information of the Internet, but also some unusual applications may be in order, for example the idea of a background video application. We are all familiar with background music, for instance. One proposal is to show a favorite video on a wall screen. It can actually be downloaded at any time, allowing you to change the atmosphere of the room according to what is desired. Another possible application would be to display paintings, including masterpieces from museums around the world. These are probable expansions of applications of telecommunications in the future. I believe that wireless technology will basically follow what fixed wire line technology will develop.

Hochfelder:

Do you have any concluding thoughts?

Tachikawa:

On this occasion of the 50th Anniversary of the IEEE, its past achievements have already been outlined. I would like to see the organization make continuous progress in the next fifty years as well, not only in technical development, but also if the IEEE could get involved in the next fifty years as well.