Fall 2007

COVER STORY The Indispensable Technology The Business of Analog The Full-Circuit Classroom Understanding Analog Analog Technology – In Brief

THE WORLD, engineers often remark, is analog.

photo by Gary Meek Associate Professor J. Stevenson Kenney focuses on analog technology at various Georgia Tech venues including the Microelectronics Research Center. (300-dpi JPEG version - 1.3MB)

Real-world phenomena – sound, images, motion, force, weight, temperature and even time itself – are captured by analog circuits, which often (but not always) hand them over to digital circuits for data manipulation and transmission.

“I’ll put it another way – what is not analog?” asks Gabriel Rincón-Mora, an associate professor in the Georgia Tech School of Electrical and Computer Engineering (ECE). “Everything in the universe is analog – continuous in time and space. It’s only natural then that all electronic systems encase their digital signal processing in an analog shell, to draw energy and to interface with the surrounding environment.”

Like the real world, analog techniques are continuous and linear. And what analog circuits mainly do is capture, condition, amplify, and/or drive real-life signals in one way or another.

Digital technology, on the other hand, processes data in discrete steps using the flexible codes we call software. Digital’s functionality is mostly limited by processing speed and by the effectiveness of the software algorithms being implemented.

A simple example of the analog concept can be found in an old-fashioned, fully mechanical watch. Like time itself, the watch’s hands move continuously around the face – unlike a digital watch, which has numbers that pop up at intervals or hands that move in steps.

A more complex example involves capturing and reproducing sound. The grooves of a vinyl record contain actual peaks and valleys that physically reproduce almost the entire spectrum of the sound recorded on them. By contrast, a digital recording selects parts of that spectrum – in a process called “sampling” – and stores them as digital code. Those pieces are then reassembled and reconstructed into an analog signal that, when driven into a speaker, resembles the original sound but doesn’t truly reproduce it.

Digital sound is very convenient. It’s easy to copy and manipulate on a computer and transmit over the Internet without losing much integrity. Today’s “digital” music players (which actually contain numerous analog elements) are small, handy and offer plenty of storage.

But digital’s piecemeal approach jettisons some of the original audio data, thereby losing content. Serious audiophiles insist on analog sound – complete with records, needles and old-fashioned tube electronics – because they believe that maintaining virtually the entire sound spectrum produces a better listening experience.

“In the digital world, you’re switching between on and off,” explains J. Alvin Connelly, professor emeritus in ECE. “The analog world is always operating in the state between on and off.”

J. Stevenson Kenney, an ECE associate professor who researches radio frequency electronics for wireless communications, explains that digital technology’s flexibility and accuracy can compensate for some of the limitations of analog circuits.

“The accuracy of analog signal processing is inherently limited by the physics of the devices,” says Kenney, who works extensively with Tech’s Microelectronics Research Center. “And that’s not true in digital – you can keep scaling and just have more and more accuracy.”

But, he adds, “I can’t put a Pentium processor in a cell phone and expect the battery to last for more than a few seconds. Analog can do high-bandwidth signal processing at much lower powers than digital.”