Monday, 22 May 2009

 

Mixed Domain Circuits and Systems [download - 1,5 Mb]


 

   

Yannis Tsividis

Department of Electrical Engineering
Columbia University
tsividis@ee.columbia.edu
 

Some of the most innovative work in science and technology today occurs at the boundaries between fields. Although opportunities within certain mature fields may be diminishing, the combination of such fields may be new, and can present a wealth of possibilities for research and development. Similar things can happen when different sub-fields or domains within a single field are combined. This talk will argue that within circuits and systems, several different domains which are normally kept separate can be combined to advantage, producing new possibilities for research and, hopefully, new practical benefits.

At the request of the organizers, the talk will include some historical information about the beginning, 30 years ago, of what are today called “mixed-signal” integrated circuits and systems. Although such systems are certainly mixed-domain, they essentially consist of cleanly digital and cleanly analog parts, joined by analog/digital interfaces. The talk will suggest that other possibilities for mixing the analog and digital domains do exist. Further, analog and digital are not the only two domains that can be mixed within circuits and systems. It will be argued that several other combinations may offer the potential for fresh research. Several examples will be offered.

In one example, the domains of linear and nonlinear operation are mixed to make possible circuits and systems that are input-output linear, yet are internally nonlinear; another example involves analog systems that are input-output time-invariant, yet are time-varying internally, thus allowing external control of internal properties without the output’s noticing. Both approaches make possible a greatly enhanced dynamic range. A different type of internally varying, externally invariant system makes possible the dynamic variation of the internal system structure without causing output disturbances. This can be taken advantage of in creating analog systems with controllable power dissipation, so that always the minimum power is being dissipated for the signal processing task at hand.

Another example involves the mixing of the digital domain and the continuous time domain, resulting in real-time digital signal processors which operate entirely without sampling; thus, no signal aliasing occurs, and the quantization error has spectral properties which are more desirable than in the classical case. In contrast to the case with classical DSPs, activity in these unclocked processors is inherently linked to activity in the input signal. Thus, if the input activity is low or absent, the power dissipation decreases accordingly.

A final example involves hybrid computation in the context of VLSI and ULSI. Today’s technology allows the implementation of a special-purpose analog computer with significant capabilities on the same chip with a digital computer, along with interfaces between the two; the result is a black box that is as accurate as a digital computer, but is faster than the latter.

The slides for this talk, including extensive references, will be available on the conference Web site.

     
     
   Tuesday, 23 May 2009

 

Systems and Networks on Chips: Challenges and Solutions [download - 9.2 Mb]


 

   

Giovanni De Micheli

Integrated Systems Center
EPF Lausanne Switzerland
giovanni.demicheli@epfl.ch
 

Micro-network design can be achieved by extending synthesis technology to cope with the particular features of micro-networks and their components, such as programmable switches, pipelined links and network interfaces to programmable cores and memories. The support for optimization of the network topology and parameters will provide competitive advantage over manual solutions.

Reliable network design will address coping with signal integrity in the physical channels while supporting reliable end-to-end data transmission. Note that signal integrity may be consciously lowered while searching for operational voltages of computational/storage units and voltage swings on links that reduce the system energy consumption.

Information encoding, packetization and routing are distinctive aspects of reliable micronetwork design. Information encoding in switches or data interfaces has shown to be effective in raising the mean time to failure (MTTF) of communication links in noisy environments. Shadow registers have been successful used to increase resiliency against soft errors, and more generally, against timing errors.

Data packetization provides us with a versatile framework to encapsulate data for reliable transmission with a low overhead. Different packetization schemes and packet routing strategies have also been addressing specific needs for micro-networks, such as the support for guaranteed throughput on specific links in combination with best-effort communication services on other links.

Overall, the design of reliable micronetworks is a challenging research area. The combination of various design objectives, such as performance, low-energy communication, and MTTF addresses some of the current needs of the design community. Moreover, the ability to synthesize and optimize reliable micro-networks provides a competitive advantage for the realization of complex multi-processor SoCs.

This keynote talk will survey the area of System on Chip design, with particular reference to the challenges provided by the 65 and 45 nanometer node technology, as well as by novel silicon (and non silicon) -based technologies. New and future design styles and methodologies will be presented, with an eye on the medium and long term technology trends.

     
     
   Wednesday, 24 May 2009

 

Foundational research of intelligent systems in the European programmes [download - 1,9 Mb]


 

   

David Guedj

European Commission
Future and Emerging Technologies
david.guedj@cec.eu.int
 

The European Union is preparing to launch in the beginning of 2015 its new framework programme for research and technological development. The programme will cover the years 2015 – 2013 and it will include all areas of science. A substantial part of it will be dedicated to information and communication technologies or ICT’s. Broadly stating, the framework programme is divided in four broad parts: ‘ideas’, ‘cooperation‘, ‘people’ and ‘capacities’. The ‘ideas’ programme is a new element. It would fund basic research to individual teams through the European Research Council (ERC) and would constitute a ‘European NSF’. The ‘cooperation’, ‘people’ and ‘capacities’ programmes would continue the legacy of the well-know EU cooperative research programmes, Marie Curie fellowship schemes and research infrastructure funding, respectively.

The basic principle underpinning European programmes is subsidiarity: the only activities that are eligible for funding are those that could not be undertaken at the member state level. In other words, there must be added value in conducting transnational European research. Typically a research consortium of a European research project has 4 to 10 partners from different member states. Participants from outside Europe can also participate, and in some cases receive EU funding. In practice the range of funding is 1 – 15 M€ for a duration of 2 – 5 years. To assess impact of the European research programmes continuous indicator collection and follow-up schemes have been set up, including commissioned studies.

An obvious area of collaboration is foundational IT research, and it is the mission of the Future and Emerging Technologies (FET) arm of the ICT programme to foster embryonic risky research that carries a potential for breakthroughs. The past and present FET activities have shown that the strengths of European research are often in the capacity to bring together multidisciplinary research teams and to create new research communities.

During 2009 the content of the new programme will be defined. Intelligent systems has been one of research topics and it will continue to be funded under the forthcoming research programme. More specifically, areas such as basic research into intelligent and cognitive, pervasive computing and communication, software intensive and complex systems as well as novel architectures are among the candidates being considered. While some parts of the ICT programme will fund research from a more conventional point of view, FET will be seeking new paradigms and approaches to tackle the hard challenges in these areas. For example, new insight into machine intelligence can be gained by studying the very basic mechanisms of intelligence as it appears in the living world, in our brains or in animal colonies.

A few examples of recent projects are presented.

     

 

 

 

 



Home | Sponsors | Call for Participation | Session Schedule | Keynote Talks | Tutorials | Special Sessions |

| Exhibitors | Exhibition Packages | Exhibition Brochure | Sponsoring Levels | RegistrationAccommodation | Hotel Information | Maps |

| Kos Island | Island hopping  | Excursions | Transportation Info | Transportation MapPresentation Info | Info for Session Chairs |

Conference Committee | Technical Program Committee | Call for Papers | Related Conferences |

| Conference Management | Contact Us |

 

 

 

Artwork and website designed by Enter Graphics(tm) and Konstantinos Kakarountas