Dumb Microprocessors and Smart Communications: A Perfect Marriage for the Future?

EECS Joint Colloquium Distinguished Lecture Series

pic of Gerhard Fettweis

Dr. Gerhard Fettweis
Professor and Mannesmann Mobilfunk Chair for Mobile Communications Systems, Dresden University of Technology
Chief Technical Officer, Systemonic, San Jose/Dresden

Wednesday, August 29, 2001
Hewlett Packard Auditorium, 306 Soda Hall
4:00-5:00 p.m.

Abstract:

Future communications will be driven by being able to connect ubiquitously, using novel smart algorithm and communication system architecture approaches. This will drive a broader vision of wireless connectivity, which shows that we are currently just at the beginning of what wireless communications will mean to us in future.

Digital Signal Processors (DSPs) are mainly focused on being applied as embedded processors in communications systems applications. They typically are manufactured for repeatedly executing the same program for the same application. Therefore, most effort in DSP architecture development so far has been in optimizing the number crunching efficiency of signal processing benchmark algorithms that are important for communications. Little effort has been paid on adding “smarts” to support demanding and run-time dependent program control, effectively making “dumb” number crunching microprocessors DSPs in the eyes of processor architects.

Communications signal processing requirements are determined by the demand for increased data bandwidth and power efficiency. In particular in the field of wireless applications the demand for network capacity per unit coverage area is addressed by new multi-antenna space-time innovations, requiring large leaps in signal processing power. However, are DSPs in terms of power consumption and die size efficiency the correct answer over hardwired solutions? Is the additional demand for standard upgrades and modifications over the air, as driven by the market demand and addressed by software defined radio research, the only request for DSPs to be used? Or is it the fact that the complexity of wireless systems results in typical chip debug cycles of hardwired chips of many years, and therefore the embedded DSP is the only way out?

The vision of ubiquitous wireless connectivity, as presented, will be enabled and driven by smart communications innovations running efficiently on DSP.

Biography

Gerhard Fettweis received his MSc/Dipl.-Ing. and Ph.D. degree in electrical engineering from the Aachen University of Technology (RWTH), Germany, in 1986 and 1990, respectively. From 1990 to 1991 he was a Visiting Scientist at the IBM Almaden Research Center in San Jose, CA, working on signal processing for disk drives. From 1991 to 1994 he was a Scientist with TCSI, Berkeley, CA, responsible for signal processor developments for mobile phones. Since September 1994 he holds the Mannesmann Mobilfunk Chair for Mobile Communications Systems at the Dresden University of Technology, Germany. He has been an elected member of the SSC Society’s Administrative Committee since 1999, and of the IEEE ComSoc Board of governors from 1998 to 2000, respectively. He has been associate editor for IEEE Transactions on Circuits and Systems II, and associate editor for IEEE Journal on Selected Areas in Communications wireless series. Over the years he has organized and been on the program committee of numerous IEEE workshops and conferences. Since 1999 he has been the CTO of Systemonic, a startup spun out of TU Dresden, focussing on broadband wireless chip-sets.

231cory@EECS.Berkeley.EDU