Compiled by: Committee for Study of Long-Term Perspective of Electronic Industry, Japan Electronic Industry Development Association (JEIDA)
Current Technologies and Future Prospects 1.Information Processing Technologies (1) Scalable computing 1:Importance of parallel processing The recent development of microprocessors (MPU) is remarkable. To meet the demand, parallel processing computer configuration with multiple MPUs is essential. The mainstream methods are SMP (Symmetrical Multiprocessor) in which memory within the same apparent distance from processors is shared and NUMA (Non Uniform Memory Access) where distances to memory are different between self processor and other processors. 2:Technical challenges of SMP and NUMA for realizing parallel processing The SMP technology allows easy programming, and the existing programs developed for a single processor can be used without any modification. Most commercial computers adopt this method. Due to the limitation of the high-speed switch linking memory and processors in the same distance, however, there is a limit as to the number of processors (scalability). Since the NUMA method allows different distances from processors to the memory, a number of processors which cannot be achieved with SMP is available with it (currently in the order of 1024 units in the message passing method). However, the existing programs developed for a single processor must be modified. In addition, for the NUMA technology, there are technological problems in many methods. To achieve some dramatic innovation in this area, development of scalable shared memory technology, scalable NUMA parallel programming technology, capacity on demand, data load technology, and I/O technology is necessary. (2) Middleware The challenging development targets for middleware can be summarized as follows. 1:Breakthrough in sophisticated reuse/ abstraction technologies and the problem of magnitude While the degree of integration of LSIs increases 40-50% annually, the software production and development increases only 5% annually. To solve the problem of magnitude, degree of abstraction for software development and the reuse rate should be increased in principle. So should also technology to accumulate highly abstract components and mass customizing to provide solutions to customers by combining these components. By handling human communication and company models, need-oriented engineering should be incorporated to produce design processes which integrate products and processes. 2:Ensuring sophisticated high quality technologies and reliability As software becomes popular and built into every area in society, social responsibilities for software as well as for the existing missions or safety-critical systems become important. Currently, technologies to ensure the quality of software are immature. A breakthrough is necessary in terms of the concept of the development process and quality itself. Formal description of specifications, proof technologies, and computation theories should be comprehensively reviewed from the standpoint of software engineering. 3:Ensuring multidimensional open technologies and relevance The technology to develop a unified platform from the technology at the mainframe level to the general computer level is promising from the standpoint of the development of role models. The operating systems themselves will become components and their customization will be possible. Integration of software and information will be one of the major problems for middleware in the future. Middleware technologies will bridge the gap between the hardware and operating system levels and provide solutions to customers. Technologies by which basically abstract systems are built up and the customers in the real world who are people with preferences, and requests and people who engage in social activities such as missions by organizations should be considered simultaneously. (3) Magnetic recording 1:Ongoing improvement of high-integration The increase in recording density on magnetic disks is remarkable. Recently, the rate of increase of surface recording density reaches 100% annually due to the GMR (Giant Magneto-Resistive) head. Around 2003 when new technologies will be introduced, it is possible that the rate of increase in surface density will be raised further. 2:Introduction of plastic substrates for cost reduction When their use in households is considered, further cost reduction is essential. The use of plastic substrates is therefore promising. There are strong demands for removable substrates, and for the development of a reliable drive structure, cartridge structure, and head/disk interface (HDI). 3:Development of next-generation technologies of vertical magnetic recording The largest breakthrough in magnetic recording in the future will be the development of the next-generation vertical magnetic recording technology. Currently, patterned vertical recording media are being investigated. Research still focuses mainly on media. Memory functions such as data write and read will be investigated and the surface density will be higher. 4:Magneto-optic recording with higher integration The technological trend of magneto-optic recording is towards the utilization of blue LD, magnetic domain expansion regeneration, and short range light. In 2005 and later, a high density of over 100Gb/in2 will be achieved. Accordingly, high-speed regeneration as well as improvement of record density will be important in the future. The floating head will be used to minimize the amount of floatation of the magnetic head. In addition, for phase change recording, a density comparable to one of magneto-optic recording will be achieved through utilization of blue LD, two-layer recording, and short range light. High-speed regeneration due to the development of new materials is also expected, which will eventually become storage suitable for both AV equipment and computers. As the resolution of the record regeneration system increases, the existing master cutting machine cannot ensure the characteristics of the original. Therefore, establishment of high-resolution lithography technology utilizing the deep UV laser beam and electron beam in the optic disk manufacturing process will be the key technology to put the future high-density optic disk to practical use. (4) Semiconductor memory 1:DRAM (a) Larger capacity Mass production of 256M DRAMs starts in 1999. In 2003 and later, 512M will be a mainstream instead of 256M. Prototyping of 1Gbit products utilizing the 0.13ƒÊm process will be conducted up to the end of 2001. The products will be made commercially available between 2002 and 2003. (b) Higher speed DRAMs will obtain higher speed to respond to the needs of manufacturers as users. This is based on an increase in the operating frequency of the CPU and base clock. The base clock frequency of the system bus will be increased from 100MHz to 133MHz. D-RDRAM, PC133-SDRAM, or DDR will be used. 2:Flash memory Asking demand for microminiature flash cards will appear as the demand for media for recording freeze-frame pictures and music will grow and that due to cellular phones will also increase. Capacity will shift from 32M to 64M. The manufacturers are preparing for the future expansion of demands. Stacked CSPs with low power SRAM and flash memory will be increasingly used. Miniaturization technology for cells of 1B bit products and technology to prevent write errors are being developed. 3:Ferro-electric memory The nano-second level has been achieved for rewrite. The maximum capacity is currently 256k bit, and this will be used for the generic memory in the distant future. This will be applied for mainly to IC cards. Currently, PZT and SBT are the mainstreams. On a product basis, the 2T(transistor)2C(capacitor) type is used. There has been a published research regarding 4M bit ferro-electric memory of the 1T1C type. 4:MRAM (Magnetic Random Access Memory) MRAMs are nonvolatile magnetic memory which have a read/write time comparable to SRAMs and a degree of integration comparable to DRAMs. Currently, the write time is 10-50ns, read time is 10-50ns, power consumption for operation is 10-400mW, and the number of available rewrites is 1015. These values are achieved by using TMR film as the material for a large MR ratio of the resistance change rate. However, there are problems in obtaining a MR ratio more than 30% and suppressing resistance variations to achieve a capacity of 1Mbit with a read/write speed of 10ns. 2.Network Technologies (1)Cable technologies 1:Optical communication will be mainstream. The transmission method most suitable for the cable technology infrastructure is by optical communication. As communications demand expand as represented by the Internet, the expansion of communications capacity of optical communication systems has been remarkable. It exceeds Moore's law, which has been dominant in the growth of the semiconductor technology. This rapid expansion has been possible due to wavelength division multiplexing (WDM). By 2010, up to 800 wavelengths will be provided by multiplexing. If the transmission capacity for each channel is 10Gb/s, a large capacity transmission system with a total transmission capacity of 8Tb/s will be possible. Such communications systems will be built regardless of public networks or LANs. Development of devices used for the systems will be an important issue. 2:Breakthrough for optical parts of higher performance Cost reduction, high performance, and miniaturization of optical parts such as optical switches, wavelength conversion LDs, as well as wave dividers and synthesizers are required. As highly integrated LSIs had a large impact on the system hardware and architecture of computers, a breakthrough in high performance of optical parts will be necessary. Any manufacturer who achieves such a breakthrough will provide the world's standard parts and systems and have a large effect on the method of system construction. For optical interconnections which are essential for large capacity communication systems, the transmission rate between boards and racks must be dramatically increased for dynamic processing of high-speed signals to achieve products with a throughput of 100Gb/s. Such a throughput results in a large-scale system with the existing technology. It must be lead to a compact size with lower power consumption. For connection of multimedia devices, high speed IEEE1394 (1Gb/s) will be achieved. 3:Technological breakthroughs to realize the next-generation Internet For nodes, many technological breakthroughs to realize giga-bit routers to adapt to the next-generation Internet are required. For high speed processing with parallel processing of switching control processors, improvement of IC elements for the switching mechanism, innovations of the address management method which makes up a large portion of router processing, and the IPV6 technology that resolves address shortage and expands usage of IP including wireless, many technological breakthroughs are required. Gateway functions to connect different networks such as IP networks and the existing telephone networks should also be developed. (2) Radio technologies 1:Rapid development of radio applications In the past 10 years, use of radio applications have expanded dramatically. In the mobile communications area, a cumulative total of subscribers for cellular phones and PHSs combined reached 50.51 million (at the end of July, 1999). This is approximately equal to the number of all households in Japan. Mobile communication terminals have been widespread so that the goal of "one for one household" is now a reality. For indoor use, high speed standards for radio LANs have been planned in offices, and various radio technologies suggested to connect PCs, portable information terminals, and AV equipment in households. Radio communications technologies are used in many fields. In the future, radio technologies will be used in many other areas such as the subscriber line radio system which use radio communication in the subscriber line between the user's household and the telecommunications line facility of the communication provider, IMT-2000 which speeds up the data transmission rate of mobile communications dramatically, MMAC which allows real-time video transmission by faster transmission rates, and satellite communication which connects the entire world seamlessly. 2:Breakthrough technologies in radio communication To utilize radio technologies in a variety of areas and develop them, some technological breakthroughs will be necessary. First, the development of a new high-speed transmission method and frequency resources (technology to use quasi-milli waves and milli waves and reservation of frequency bands) is necessary to use radio communications as the transmission infrastructure of multimedia information. Then, researches regarding sophisticated wireless network configuration such as development of mutual communication technology such as wireless LAN, mobile public networks, optical high-speed communication networks and researches of QoS in wireless TCP to realize real-time information (video and voice) in radio communications. For wireless terminals, direct control of radio with software will be possible such as software radio in which a single terminal adapts to multiple radio methods in terms of software. Research on sophisticated wireless terminals will thus be required. (3) Information Distribution As important technologies in information distribution in the multimedia era where communication and broadcasting will be integrated, there are the QoS (Quality-of-Service) technology of the IP network and the information compression coding technology of multimedia contents containing dynamic images and voices as represented by broadcast videos (MPEG). 1:QoS technology for the IP network The protocol RSVP (resource ReSerVation Protocol) which reserves the end to end band along the path will be standardized by 2003, and multicast-compatible RSVP will be standardized by 2005 and go into practical use matched to the hierarchal coding of contents. In multicast, the resending mechanism of recovery processing will be much more complicated compared to unicast, and there is the challenging prospect of reliable multicast for which a high quality is required. The protocol WAP (Wireless Application Protocol) for radio networks will be standardized in turn and go into practical use between 1999 and 2002. The protocol COPS (Common Open Policy Service Protocol) and DIAMETER for controlling networks in a unified way will be standardized around 2001 after going into practical use first. Other new protocols may be suggested. 2:MPEG technologies The series of development and standardization activities from MPEG-1 to MPEG-4 regarding the information compression coding technology for multimedia, the technologies which cover a wide range of targets and permit many applications have been established. Since MPEG-4 covered the compression coding in low bit rates, the development of technologies for compression coding of video seems to have become stabilized. Information compression is compression of information content to be transmitted. It is natural that technologies that utilize the transmission lines for providing contents and extract desired information from a large amount of contents and handle it at will become centers of interest, rather than seeking the limit of compression. The future theme will shift from "information compression" to "information organization". MPEG-7, which is currently under review, tries to define a framework for structured description of information regarding contents to facilitate distribution and usage of multimedia contents. The objective is to standardize description of added information, attributes, and characteristics of contents for easy search and categorization. It is expected that this will foster utilization and widespread use of contents which in turn will promote the development of applications. The problem of protection of intellectual property right related to this should be resolved. (4) Security 1:Necessity for quantum encryption technology Improvement of performance of computers results in a situation that ciphers widely used in the world may be easily decoded. To strengthen the competitiveness of companies and for national security, the continuous development of cipher systems will be required. Quantum cipher utilizing the physical principle of quantum mechanics unconditionally ensures safety against tapping. Research which is influential in terms of technology development and exploitation of new applications is expected. 2:Necessity for electronic verification technology For transactions on networks where other parties are invisible, the electronic authentication function which utilizes cipher technology is essential for verification of the identity of the partner and assurance of integrity of communication data. Today, the method utilizing digital signature based on public key cipher is the mainstream. However, this does not create a solid awareness of performing a legally binding action as distinct from manual signature and imprint. Development of an electronic signature system which is easy to use and implements the person's intention reliably is needed. In connection with this, operations have begun to stipulate electronic signature and authentication methods in Japan. In terms of the system, improvement of environment regarding electronic authentication is rapidly done. 3:Establishment of the Public Key Infrastructure In order for electronic authentication to be widely used in the future, a public key infrastructure (PKI) will be necessary. This becomes the common core for various information security systems. The demand for authentication services utilizing PKI is increasing in the private sectors. As a result of policies to realize electronic government, electronic application and electronic procurement will be widely used by national agencies and local governments. Development and extension of PKI in a wider range will be necessary. 4:Development of security technology for IC cards IC cards are also important as an element for security. Development of technologies such as integrating the next-generation IC cards into PKI as one component and incorporation of functions that recognize, record, and match living-person characteristics data to identify the cardholder are expected. It is important to accelerate development of LSIs most suitable to IC cards and to resolve problems of power saving for non-contact IC cards. 5:Establishment of network security technology allowing access control by user or by device For networks, measures against illegal accesses by hackers and computer viruses are important issues. As information electrical appliances and portable information terminals become more popular, the connection of all devices as resources of the network as well as the existing connection via desktop PCs will become common. Thus, the development of network security technology allowing access control by user or by device such as incorporated VPN (Virtual Private Network) will be an issue. In addition, the use of the mobile Internet such as the i-mode service becomes popular and the assurance of security in the radio environment is important. 3.Usability Technologies (1) Input/output: Knowledge acquisition As digitization of various types of information advances, creation of attractive contents and services, standards for standardized, safe, and easy to use information distribution, and implementation of information communication platforms enabling the use of information services in a unified way will be needed. The following breakthroughs will be necessary to adapt to an increase of information traffic, added values of information due to digitization, integration of information flow and distribution, and changes in lifestyle due to the global network. 1:Implementation of natural interfaces For widespread use of multimedia, development of natural interface technology which will realize integration of operation systems, easy operation, integration of application interfaces, seamless and scalable connection to peripheral devices with ease of use is important. 2:Establishment of agent technology Agent technology that supports human information activities such as intelligent search which realizes intelligent information access, communication support, and inspiration support will be important, and its develop should be fostered and accelerated. 3:Substantial digital archives Digitization of multimedia such as images, videos, and documents will progress. Today, due to the development of XML, computerization of knowledge resources can be realized. As the next step, it will be an urgent priority to turn information resources into knowledge resources to utilize them. The digitized multimedia information should be collected via information extraction technology such as text mining, saved as digital archives, and made readily available. 4:Development of automatic translation software To use resources overseas, or broadcast resources to overseas, automatic translation software for multiple languages should be developed. In addition, language information archives (language information dictionaries) should be developed. 5:Establishment of language recognition technology As accuracy of voice recognition has been improved, construction of systems applying the high-precision voice recognition technology is being pursued. The key technology to promote this will be the integration with technology to understand meaning of statements in foreign languages. 6:Necessity of development of environment (infrastructure/system) o For standardization of the agent communication protocol, the system to realize standardization (including participation and contribution to international standardization) of the technology in collaboration with the private sector, government research institutes, and administration should be developed by 2002. o An incentive for SOHO (satellite office) should be developed by 2002 to promote SOHO. o For local promotion of information archives, a social infrastructure investment plan to foster local development of information archives (electronic libraries and electronic museums) should be developed, and investment should be fostered. (2) Input: Voice and image recognition 1:Fostering of development of universal and generic interfaces Technologies regarding so called natural interfaces, which apply the means of communication between humans to communication between humans and computers, have been rapidly advancing along with improvements in the processing performance of computers. Technologies such as character recognition, voice recognition, and individual identification have been put into practical use. In addition, some research regarding operation input with arms, legs, and eyes, and detection of brainwaves have been started. However, even advanced technologies such as voice recognition which have been put into practical use are used under ideal conditions. However, they assume users who are familiar with operation of computers. In the future, development of more universal and general technologies for transparent interfaces which do not depend on a specific environment and are operable without special training will be important. 2:Fostering of development of compound technologies with output and intelligent recognition Input is not only an input operation but a part of a interactive, two-way interaction. The important challenges in the future will be the development of total interaction technology which involves responses (outputs) to the input not restricted to input technology, support for communication in addition to input and output such as language translation technology, and intelligent recognition technology which guesses the meaning from the context. (3) Display (output/graphics) 1:Potential of plastic film substrates In the 21st century, all familiar devices will have some kind of computer functions as a result of development of digital and network technology against the backdrop of large capacity communication technology. Output technology which consists of a human interface will be important. When high-resolution, high-speed, large, and low-cost displays which can adapt to all graphic representations are needed, processors, peripherals and display may be mounted on the same substrate. Plastic film may be used as a substrate, as well as glass. A further promising substrate candidate is the use of paper displays or systems on paper, which can be rolled but not folded. This would be the ultimate form of assembly technology, with Japan in a leading position in the display manufacture technology. 2:Speech synthesis technology Future development of speech synthesis technology as output is expected. We vibrate our throat and move our mouth and tongue when we speak. A more user-friendly speech interaction system can be realized by modeling the information processing mechanism which implements human speech and by replicating it on the computer. Importance of the five-sense output devices We have highly sensitive sensors such as tactile, olfactory, and gustatory senses. When we sense phenomena in the outside world, we use some or all of the senses and perform information processing on a major scale. If such five-sense information can be encoded precisely, transmitted from a distant location, and decoded, the computer society will make a quantum leap. In the future, how to realize five-sense output devices would be a research subject. 4:Expanded reality technology as next-generation display technology The 3D graphics technology will advance in the future. Graphics technology to superimpose the real world over the 3D virtual reality will be developed. This technology reinforces and enhances the real world electronically, so it is called "expanded reality" or "enhanced reality". Virtual objects can be superimposed over the view of the real world through glasses by utilizing a see through head mount display. Similarly to the subtitles of movies and television programs, characters and figures can be presented, or supplemental remarks, figures, and procedures can be displayed according to the situation. There will be many applications for this. 5:Problems of software in the expanded reality technology The technology to recognize the real world and to superimpose the virtual world over it is very sophisticated, and this should be pursued as a new technology area. The eyes of computers which recognize 3D should be a system that understands the meaning of the objects, learns, and automatically superimposes virtual data the user needs over the real world, involving not just pattern recognition. Extensive basic research should be done regarding the borders between the image/voice space and the real space in production of 3D media because the observer would tell if it is the real 3D space or not. The problems of the observer's mental responses (nausea and motion sickness) and asthenopia due to sudden changes of objects. Research on specific applications will be important. Displays should always allow for a comfortable display environment by feedback of the user's responses to the system. Since responses may vary individually, and there are many applications, an automatically-developing mechanism which has a self learning function will be needed. 4.Other Technologies (1) Micromachining Micromachining has been reducing the device cost 25-30% annually on the average per function. Micromachining technology has contributed to high performance of applications and cost reduction. However, it is approaching the physical limit, and a technological breakthrough is necessary before 2010. 1:Breakthrough of processing limits of lithography Lithography technology in the range of visible light has been used as micromachining technology for semiconductors since its invention. Lithography technology with visible light is of the order, at best, of 100nm, and the next candidate technology should be developed promptly. EUV, EB direct writing (EBDW), SCALPEL, 157nm DUV, proximity X-ray, and ion beams would be candidates, but their ranges are where the existing masks, resists, lenses, and optical systems cannot be used. A technological breakthrough is needed. 2:Development of new device architecture If device technologies and system architectures develop as extensions of the current ones, the device speed will reach the physical limit shortly. There will be no innovation without a large change in the architecture and basic device structure. 3:Establishment of measurement technology Since exposure to light has reached the limit, measurement of more microscopic parts will be more difficult than exposure. Some radical innovation should be sought by combining technologies the reaching physical limits and including statistical processing. (2) Power management (secondary battery) Secondary batteries are the key components of mobile solutions. To realize paper computing, which is an ideal form for the mobile PC, innovation along this trend would be necessary. To realize electric cars and hybrid engine cars early, development of large lithium batteries which are small, light, and safe is necessary. Therefore, development in this field should be fostered. Fostering of development of polymer-type batteries, promising to realize thin and light secondary batteries, is essential for development of technologies for mobile solutions. (C)Copyright,JEITA,2000
|