Written by: Bernt Ericson

Applied research
The role of research at Ericsson is to investigate new ideas, in order to understand their usefulness as applied to telecommunications. Promising ideas are prototyped to the extent that the development organization can integrate new findings into commercial products. Applied research of this kind relies heavily on input from the research community at universities and research institutes. Academic research is often accused of being introverted and of little benefit to society. There are, however, many examples of how the needs of society and industry have inspired the basic research corps to expand theories into new fields—new-found knowledge gives rise to new products which inspire new research ideas, and so on.

I was recently invited to speak on the relevance of basic science at the “World Conference on Science” in Budapest. In my speech, entitled “Views from the electronics industry,” I explained why and how cooperation between industry and academia can benefit each party. I began my speech with some historic highlights that have played a fundamental role in the evolution of telecommunications and the information society.

Historical highlights
Thomas Alva Edison was obsessed with the idea of turning electricity into light. After thousands of experiments with various materials, he succeeded, in 1880, when he perceived that the glowing carbon wire must be protected from oxygen in the air. Scientists who analyzed the light bulb found new phenomena that led to the discovery of the electron, in 1897. This addition to the pool of basic science led to the development of the diode, and later, in 1907, to the triode—the triode is needed to build an amplifier, which is an essential component in telecommunications.

Alexander Graham Bell filed the basic patent on telephony in 1876 (the same year in which Lars Magnus Ericsson founded Telefonaktiebolaget LM Ericsson). This new form of communication was as instant “hit” and quickly grew in use. Initially, the distances were bridged by open-air copper wires. Mother Nature, however, set a limit on how great a distance could be traversed (as explained by Ohm’s law). Early trials to traverse the Atlantic Ocean failed because the required voltage was so great that the cables quickly broke. With the invention of the amplifier, however, it was possible to connect New York to San Francisco, in 1915, and at long last, in 1956, to inaugurate the first successful transatlantic cable. The triode also opened the era of electronic computers, although to build a practical machine, a veritable host of tubes was required. Not surprisingly, power consumption and heat dissipation were major issues, but the greatest limiting factor for operating and maintaining early computers was the high failure rate of individual tubes. The US military (which was quick to see the potential of automated computing) strongly urged and supported a better amplifying component. In 1949, the transistor was born when researchers at Bell labs found the answer they were searching for in semiconducting materials. Since then, engineers have steadily found new ways of integrating more and more transistors into the same silicon chip (a 400% increase every third year—an evolution that follows what has come to be known as Moore’s law).

Moore’s law
The integrated circuit plays a key role in most of Ericsson’s products. Thanks to constant increases in performance, we have, at regular intervals, been able to release greater computing power in AXE. For instance, in this issue of Ericsson Review, you can read about the latest APZ 212 30 processor, which contains several highly integrated custom-designed circuits (the largest circuit is composed of 10 million transistors!). In telecommunications, the need for greater capacity can to a large extent be traced to the rapidly growing use of wireless communication. As allocated frequency bands become saturated, new methods must be found that make optimum use of these sparse frequencies. The adaptive antenna (also discussed in this issue) is an example of one such method. Initially, the theory and first practical implementations of this technology were driven by the needs of the military. Today, the technology has matured and can be deployed in civilian systems. According to Moore’s law, we can expect increases in hardware performance that will allow most new functionality to be added through software. We must therefore have early access to, and become very adept at applying new technology, especially as relates to sotware. Consequently, we seek to establish good cooperation with academia and are constantly on the lookout for breakthrough research.

[First published in Ericsson Review no. 03, 1999]