Technology and Human Thought Drive Future Communications Systems

December 1999
By Sharon Berry

Problem-solving techniques grow with electronics advances, but new riddles emerge to vex planners.

Rapidly evolving communications techniques are leading scientists to integrate technology trends and human methods of thinking to solve problems that are yet to be encountered.

Experts at many high-technology firms have identified keys that will have far-reaching effects on thousands of future devices. Among the notable devices are handheld video telephones that will change how public safety personnel work, 64-bit processors capable of speeds of 800 megahertz that will accelerate the development of speech recognition interfaces, and personality-bearing intelligent computer agents that will automate more routine business and household tasks.

Moore’s Law portends a doubling in semiconductor processing power, or circuit density, every 18 months. This phenomenon already has had a tremendous impact on communications and handheld devices. According to Raju Rishi, director of international information technology solutions, Lucent Technologies, Holmdel, New Jersey, four fundamental concepts will change the way the general population communicates in the future.

Miniaturization is the first concept. The density- doubling trends indicate that a device such as a cellular telephone will become a quarter of its size every three years. Early in the next century, cellular telephones could be the size of a large coin and would be mainstream technology.

The ramifications of reducing the size of commonly used items are significant. For example, a cellular telephone that is the size of a quarter will require new and different user interfaces because a dial pad will not be an acceptable user interface for a device of this size.

Miniaturization will also allow 360-degree video cameras to be incorporated onto a single chip, created through a concept known as triangulation. Digital zoom technology will be the result. Triangulation is a technique by which three or more cameras simultaneously take still or moving pictures of an object, which can then be rendered in three dimensions. Digital zoom technology uses a computer to then make educated guesses about how an image would look if it were magnified.

Microelectromechanical systems (MEMS) technology will to be an important part of the miniaturization work. It will allow devices to be placed on a single chip by etching moving parts into silicon with processes that exist today. “One of the things you see about a cell phone today is that the microphone and the speaker tend to be independent from the circuit board of the cell phone. We’re finding that we can actually incorporate microphones and speakers onto the chip itself, thereby reducing the size considerably,” Rishi shares. Because a speaker requires moving parts, MEMS technology becomes valuable.

The second concept that Rishi believes will revolutionize the future of communications is advances in user interfaces. When the sizes of telephony and PC components are reduced, some of the traditional interfaces such as a dial pad or keyboard can no longer be used effectively. Consequently, technologies such as speech recognition are accelerated and driven by miniaturization.

Technologies that are enabling speech recognition are very important, Rishi declares. “One begins to question why speech recognition hasn’t already taken off, but I think in this next millennium speech recognition is going to be one of the foremost technologies that we are going to be focusing on,” he offers.

Speech recognition requires fast processing power, and Moore’s Law is important for it to be realized. This idea also requires very large storage capacity hard drives. The capability to store names and vocabulary combined with the tremendous increase in processing power are key to enabling speech recognition applications.

This user interface enhancement will lead to intelligent computer agents that can simplify automated processes. The agents will also manage the overabundance of data. Businesses as well as individuals could offload routine tasks to intelligent agents. “You might say, ‘Agent, come up!’ and you’d hear a voice response saying, ‘Yes ... what would you like?’” Rishi notes.

In his book titled The Age of Spiritual Machines, Ray Kurzweil foresees that these intelligent agents or assistants will be commonplace by 2009. “Intelligent agents that combine continuous speech recognition, natural language understanding, problem solving and animated personalities [will] routinely assist with finding information, answering questions, and conducting transactions,” Kurzweil predicts. These assistants also will come in a wide range of personality choices. Consequently, personality designers will be in demand, which will spur a growth in this arena of software development.

According to Rishi, the third concept that will revolutionize communications technology is the elimination of the obstacles posed by national boundaries. Today, the Internet enables sales from one location to another location globally. So, for example, a person could find the least expensive price of a product and have the item shipped even if it is offered by a company in a country half a world away. This is decreasing limitations, Rishi explains, not only in terms of sales and service value, but also in terms of resources and staffing. “Even on our projects, we’ve got people in multiple countries collaborating on single systems. That’s a very powerful issue because it brings to the forefront new concepts of how you deal with different cultures, but it also gives you access to a plethora of information and sources of service,” he adds.

Ubiquitous communications is the fourth driving concept. However, the challenge of achieving this is how to enable different configurations of devices to talk to each other. “The hurdle is a requirement of some intelligence in the network or in the endpoint that says, ‘I am going to determine the level of sophistication that each endpoint has’ and then makes the provisions to allow that call to happen,” Rishi concludes.

Dr. Randall Isaac, vice president for systems, technology and science, IBM Corporation, Yorktown Heights, New York, views the future of technology and how to approach it in a different light. Isaac contends that technology is falling into place around the banner of e-business, and research is being focused accordingly. This includes a transformation of business processes.

Certain questions should be addressed when conducting research for future technologies, Isaac recommends. How do you revolutionize the business process with new ways of communicating? How do you then write the software and create applications around the improved process? Another consideration is helping customers. “We think that, as we move into the 21st century, many users are going to need to build their own applications quickly and reflect their own personalities and their own approaches,” he notes.

Isaac maintains research must focus on topics such as running an e-business from a hardware context as well as translating information from computer-readable to human-readable format. These dramatic changes and solutions are what enable e-business to put customers in a competitive position.

He suggests four main fields that will be essential to the future of communications and technology. The first is the fundamental silicon processor. It will continue to progress at the current pace for the next four to five years, but toward 2010 some problems will arise in lithography and defining patterns.

Gordon Moore, creator of Moore’s Law, identified the underlying driving force behind these trends. “He pointed out that 50 percent of the effect [of chip progress] was due to lithography alone, and all the other long list of factors combined make up the other 50 percent,” Isaac explains.

Semiconductor manufacturers use optical lithography in a highly specialized printing process used to put detailed patterns onto silicon wafers. An image containing the desired pattern is projected onto the silicon wafer, which is coated with a very thin layer of photosensitive material called resist. After development, the resist forms a stenciled pattern across the wafer surface that matches the pattern of a circuit. Manufacturers already are able to place millions of transistors per square centimeter.

Last year, researchers started patterning features smaller than the wavelength of light being used to set the pattern. However, as features get smaller, peculiar optical effects arise. Isaac projects that in five to 10 years, the normal light currently being used for lithography will hit a dead end. New sources such as X-ray, scalpel and extreme ultraviolet will need to be explored.

Lithography today uses a deep ultraviolet light source. According to Howard High, corporate communications manager, Intel Corporation, Santa Clara, California, “It’s like when you use a writing instrument on a piece of paper. If I give you one of those fat crayons for 4-year-olds, it is very hard to make fine lines. In etching fine lines on a chip, it’s the same thing, you need a very narrow wavelength of light.”

The three additional fields in communications identified by Isaac are storage, data transfer and displays. In storage, magnetic drives are doubling in capacity every 18 months. Isaac believes that by 2005, designers will hit a superparamagnetic limit. This means that the bits stored on the magnetic film on the hard disk are so small they will be thermally unstable and slip around. Replacement technologies being explored include holography, scanning tunneling microscope techniques, and near-field optical imaging.

As data transfer technologies develop and fees drop, the technologies shortly thereafter will be rolled out to every home, including low-cost wireless devices and optical fiber backbones that will handle an enormous amount of traffic, radically changing communication. One of the keys to this revolution, Isaac relates, is silicon-germanium technology. Combining silicon and germanium enhances communications because the combination forms the basis of very-high-speed transistors that can attain switching speeds beyond traditional semiconductor capabilities.

Communications display is the last field Isaac notes will affect future technology. “There’s an enormous amount of computing power already. It’s a question of really getting to a low-cost, effective display that can relieve the PC from the millstone of the CRT [cathode ray tube],” Isaac offers. He predicts that flat panel displays will prevail in the years to come with competitive costs. Additionally, new display formats will liberate PC designers, enabling them to embed a screen in appliances or other objects.

With these considerations, Intel has been working on the Itanium processor, formerly known as Merced, which will include three levels of on-chip cache, with the level three cache offered in a 2-megabit or 4-megabit package. IA32 processors, like the Pentium II and III, have only two levels of on-chip cache. According to Ron Curry, marketing director, IA64 processor division, Intel, the Itanium allows the designer to start with a clean slate to consider these issues. The results will be headroom to expand and the ability to build in features that allow three-dimensional graphics performance and integration of emerging e-commerce applications.

Curry emphasizes that researchers must also examine what is still possible for a more effective technology roadmap to be drawn. Industry must consider consumers’ impending desires rather than what they currently want.  A few years ago, for example, it was not yet clear that the Internet would develop as rapidly as it has. Researchers had to look forward and recognize that, as servers, networks and computer systems would increasingly be used for business applications, security measures would be necessary.

Two additional technologies are expected to deploy in early form in 2001. The Horizon prototype is a part of the new third-generation technology, Steven Levine, manager, multimedia communications research laboratory, Motorola Labs, Schaumburg, Illinois, explains. Third-generation technology, or broadband, follows the first generation, characterized by analog technology, and the second generation, distinguished by digital technology. Motorola’s Horizon prototype unit is 5 inches high and features a camera that rotates on top. The unit will allow video operation over wireless channels and enable users to see more than just each other’s faces.

Using motion picture experts group (MPEG) technology, researchers are working toward improved video compression rates for a video telephone that transmits 165 million bits per second. In video compression, the bit stream enters a video compression application-specific integrated circuit where it is compressed to between 64 kilobits and 128 kilobits per second. Following an International Standards Organization regulation called MPEG4, the video telephone will be error resilient so the user will not lose multiple frames of information. The new standard will allow the unit to recover or conceal an error, Levine relates.

This technology has numerous advantages. From a public safety standpoint, firefighters could use the video telephone to send images of the scene back to the central office where decision makers could gauge what equipment and personnel to send out. An additional application would be personal safety. A person walking down the street could carry the device and record and transmit the image of anyone approaching. The Horizon prototype has a 2-inch display screen that allows the user to view both incoming and outgoing transmissions.

Motorola is also working on a multimedia television technology. Using MPEG4, the device would enhance the user’s viewing experience by making it interactive. Levine reports that a user could move a cursor to an object on the television picture, then highlight it to request more information. For example, viewers of a golf match could highlight a player and find out the player’s statistics. Or, from a marketing end, a user could highlight the golf ball and then proceed to an Internet site and purchase that brand.

Although the responsibility of who would program the endless information has not yet been determined, Levine sees the task as a great revenue opportunity. The television show itself would be a commercial. Therefore, it would be in the best interest of both the broadcaster and the advertiser to work together to supply and program the information. The technology also will allow the viewer to receive more than one channel simultaneously and arrange the images on the screen. This would include a transmission from another unit such as a camera in a baby’s nursery.

Leaping to 2099, Kurzweil notes that human thinking will be merged with the world of machine intelligence and communication. He predicts a new computer-self-generated increase in technology with artificial intelligence directing its own path. Although a long way off, many of the base technologies already are in place. The question then arises, Kurzweil observes, what is human?