23-10-07

SDR in Belgium

For those who are interested I hereby post some links of IMEC websites related to SDR (Software Defined Radio). SDR is the hottest topic on earth (so also in the air :-) ) at the moment.

 Software-defined radio front-ends

http://www.imec.be/wwwinter/mediacenter/en/SR2006/681523....

 Software defined radio flexible air interface

http://www.imec.be/wwwinter/mediacenter/en/SR2006/681524....

 Enjoy the readings!

 

Feel free to post links that are related to SDR as a comment in this post... I am searching to learn on this topic.

 

17:34 Gepost door Mobile blogger in Algemeen | Permalink | Commentaren (0) | Tags: sdr, rf, imec, software-defined radio, air, interface |  Facebook |

06-04-07

Bad periode for european fabs

It seems like it is going worse everyday with the European semiconductor industry.
We have some high-tech research facilities (IMEC, Crolles,...) but we can't produce or own chips in our continent?
Why is this? I don't know... But at the end of the story all the companies that are outsourcing their IC production
are not healthy because the market is like everyone does the same thing.
When you want to be a big good player you must have your own production, engineering and R&D.
When you outsource your production it can be that soon the factory that makes your product comes with the same product with another name 
 
Read the following related story about NXP outsourcing it's production of IC's to TSMC
 

NXP going fabless after 90nm.

NXP has now said that all its production will be sourced from TSMC for chips with more advanced processes than 90nm.

NXP's rationale for the move is that TSMC can provide processes better than 90nm so why go to all the trouble of developing your own?

Funny how quickly attitudes can change. When NXP was part of Philips, when Pasquale Pistorio ran STMicroelectronics, before Freescale was bought by private equity funds, it was held as axiomatic by all three companies that it was an important competitive advantage to have access to the most advanced digital CMOS processes, and that the only way to achieve that was by developing them yourself

That was the rationale for getting the EU to pour hundreds of millions of Euros into Crolles over the years. Now, all three companies have pulled out of doing basic advanced digital CMOS research at Crolles.

Now, it seems, it doesn't matter whether you develop basic advanced CMOS process technology or not. Instead you can go to a public foundry, which is open to all, and have your chips made on the same terms as everyone else.

"If every company goes to foundries, using the same processes and cell libraries as everyone else, what differentiation can a company offer customers?", asks Malcolm Penn, CEO of analysts Future Horizons, "five years down the road, when ST (and for ST read NXP as well) asks: 'Why haven't we got any customers', the reply will be: 'We can get this from loads of other people cheaper, now that they're making them in the Congo

Source:

http://www.electronicsweekly.com/blogs/david-manners-semi...

27-03-07

R&D in Europe is not healthy

R&D paradigm shifting

An ongoing debate in Europe concerns the creation of the European Institute of Technology (EIT).

The EIT, a consolidated mega research center that would house both academic and industry projects, keep talent in the EU, and unite researchers in the 27 member states under the EU flag, may well end up being a reality in Brussels. Yet the EIT proposal from the European Commission continues to go through revisions, and the concept has many critics. All of this points to a bigger issue: Europe is headed toward an R&D crisis.

Europe has long been second in annual R&D spending to the leader, the United States, and in some sectors has fallen behind third-place Japan. Now Europe is facing new and strong forces in Asia when it comes to R&D.

According to a 2006 study on global R&D by the Battelle Memorial Institute, China has increased R&D spending annually about 17% over a 12-year period, compared to 4% to 5% for Europe and the United States. As China and India pour money into R&D, offer their cheap skilled labor, and grow their vibrant economies, both countries could move into the top tier of R&D spenders worldwide.

Moreover, R&D investment in Asia is accelerating. The Battelle report surveyed global businesses and asked if the companies planned to increase R&D in Europe in 2007. About 28% said yes, and 48% said no. When asked if they planned to increase R&D in Asia, the response was 65% yes, and 10% no.

"Now we recognize that China and India are challenging us, and that's the main reason this discussion has become much more intensive," says Esko Aho, former prime minister of Finland, who last year released an influential report on innovation in Europe.

"Europe is moving too slowly, and this can lead to a crisis when we are hit by a demographic revolution and its consequences."

At first glance, an R&D crisis seems strange. Europe has many world-class research institutes: IMEC in Belgium, Fraunhofer in Germany, and VTT in Finland, to name only a few. Livio Baldi, director of R&D cooperative programs for STMicroelectronics, calls it the European paradox: lots of world-class research pockets but a poor track record of commercializing the results.

"Europe has a lot of publications but very few patents and startups," he says.

Part of the problem is fragmentation. Pockets of research are not linked to a central hub, something the EIT is intended to resolve. Another issue has been talk instead of action. Europe's "Lisbon Strategy," a grand development plan launched in 2000, promised to reshape the EU into the "most competitive and dynamic knowledge-based economy in the world" by 2010. Targets for increasing R&D spending in member states were set but not met.

At a deeper level, cultural issues are to blame. Europe's risk-averse culture limits entrepreneurial drive, and organizational structures work to lock people in their place.

"We have no tradition of people moving from academia to government to business and back," Aho says. "Innovation required that type of mobility."

Aho says he believes that the proposed EIT could be a small part of a more comprehensive solution. Among the recommendations in his report was fundamental cultural change that would encourage mobility and promote more entrepreneurial risk taking.

Jules Duga, coauthor of the Battelle report, says that what is happening in Europe is merely early signs of a gradual but enormous rearrangement in global technology dominance due to the rise of Asian giants.

"Countries will have to determine their strengths, rearrange resources, and adjust over a period of time to a changing position in R&D on a global basis," Duga says. "Expect a shift in the overall R&D paradigm."

Source: Battelle Memorial Institute and R&D Magazine’s 2007 Global R&D Report

21-03-07

Success with IMEC and Synplicity's Synplify® Premier Software

 
 

 


Success with IMEC and Synplicity's Synplify® Premier Software
Click here to learn more about the Synplify Premier tool.

 
IMEC, a European nanoelectronics research institution, used Synplify Premier software from Synplicity to demonstrate that its C-programmable reconfigurable processor architecture ADRES is feasible for use in portable wireless multimedia devices. The entire processor system was successfully prototyped for a multimedia ADRES processor instance on a Xilinx Virtex-4 FPGA through use of the Synplify Premier tool. The Synplify Premier product provided excellent support for achieving the required clock frequency. IMEC credits the Synplify Premier tool's built-in knowledge of the FPGA's physical characteristics for the accurate timing results that it delivers. The ADRES prototype system has been important for IMEC in showing that the ADRES processor architectural template and its corresponding C-compiler are sufficiently stable for use in portable devices. 
 
 
IMEC's ADRES Innovation Promises a New Future for Hand-held Multimedia Devices
 
IMEC of Leuven, Belgium is one of the world's leading independent research institutions in nanoelectronics and nanotechnology. Its research focuses on next generation chips and systems, and bridges the gap between university research and technology development in industry. IMEC's blend of know-how and corporate relationships position the organization to help shape key technologies for future systems. 
 
ADRES (Architecture for Dynamically Reconfigurable Embedded Systems) contains two views which are tightly coupled: an array of processing elements that runs the data flow part of the application and a VLIW that executes the control. For hand-held multimedia devices, this technology delivers enormous flexibility benefits over fixed ASICs because various video codec standards can be quickly and easily accommodated through C programming. In addition, ADRES-based processors offer power efficiencies six to twelve times higher than state-of-the-art C-programmed processors. 
 
With the demonstration IMEC has proven that processors based on the ADRES architecture can deliver sufficient performance. The multimedia ADRES processor instance was developed to support MPEG-2, MPEG-4 and H.264/AVC video decoding at resolutions ranging from QVGA up to D1. The demonstration employed the HAPS-32 from HARDI Electronics, which contains two Xilinx Virtex-4 LX200 FPGAs, as its prototyping board. IMEC constrained the FPGA clock input to 50 MHz to decode 30 frames/sec of H.264/AVC content at CIF resolution. 
 
Synplify Premier Tool Delivers the Necessary Performance 
 
IMEC began by synthesizing the design using the Synplify Pro product from Synplicity, the tool that had served the organization well for many years. Synplify Pro software came close to the goal at 46 MHz, but not close enough. 
 
"It was essential that we find a way to reach 50 MHz, and so we performed an investigation of the state of the art in FPGA synthesis," said Maryse Wouters, Activity Leader of the Integration Team. "Fortunately we found our answer, the Synplify Premier solution, which is capable of delivering the performance we needed. In fact it did even better than we had hoped, 52.6 MHz. Everyone was pleased with the performance gain."
 
"The reason why the Synplify Premier tool does the job better is that it understands the physical characteristics of the FPGA in fine detail and uses that knowledge to craft an optimal design," explained Wouters. "That's particularly important with the most advanced FPGAs on the market." 
 
Building on Synplify Pro technology, the Synplify Premier solution embodies its knowledge of an FPGA's specifics through a patented Synplicity technique called graph-based physical synthesis, which represents an FPGA's pre-existing wires, switches, and placement sites as a detailed routing resource graph. Graph-based physical synthesis produces rapid timing closure by automatically outputting timing-correlated legal placement and by considering availability of actual FPGA routing resources when measuring delays, rather than just physical proximity of instances. Unlike ASICs, in an FPGA physical proximity does not always correlate to timing delays, making ASIC-style physical synthesis approaches inaccurate when applied to FPGAs. Only graph-based physical synthesis can accurately estimate timing delays when performing physical synthesis.   
 
Graph-based physical synthesis also cut place-and-route runtimes significantly for IMEC. The total elapsed time for placement and routing was six hours with the Synplify Premier solution versus seventeen hours with the Synplify Pro tool. The reason is that in addition to performing synthesis, the Synplify Premier product actually places the design in a manner known to meet timing, and delivers a design that will be fully routable using the Xilinx ISE toolset. 
 
The correlation between the Synplify Premier solution's performance predictions and actuals was much better than IMEC had seen. The new tool predicted 51 MHz performance, which was very close to the actual result of 52.6 MHz.
 
IMEC's New Standard for Synthesizing 90 nm FPGAs and Below
 
With its flexibility to incorporate multiple video codec standards, the short time-to-market made possible by its high level language programmability, and its power efficiency, ADRES promises to play a major role in the next generation of mobile multimedia platforms. 
 
"Using an FPGA-based prototype platform, IMEC has demonstrated its C-programmable multimedia ADRES processor instance for real time H.264/AVC video decoding," said Wouters. "The performance gain that the Synplify Premier solution delivered was as promised in the Synplify Premier data sheet." 
 
Because of the excellent results it delivers, the Synplify Premier product has now become part of the tool flow at IMEC for future projects using leading edge FPGAs. "It is clear that for 90 nm FPGAs and beyond, the timing closure offered by the Synplify Premier tool is crucial," Wouters concluded. 
 
Click here to learn more about the Synplify Premier tool.

23-01-07

Disaster for European High-Tech

I don't feel good about the electronic industry in Europe at the moment, there are to few companies left that are pure European. It is time to work for the future, to invest in jobs... not like we are doing at this time we have to start up our economic rocket engine...

 

If we keep doing like this the semiconductor industry in Europe will be soon death.

 

Disaster for European High-Tech

It is a major blow for European high-tech as Crolles2 unravels with first NXP, and now Freescale, pulling out, leaving STMicroelectronics on its own with TSMC as a kind of junior partner.

Without Crolles 2 as a beacon of EU-subsidised process technology R&D, a lot of European work on related technologies, such as production equipment, will be de-emphasised.

Jobs will go at Crolles, and those are the jobs of highly skilled scientists and technologists which Europe should be doing its best to increase, not diminish.

OK, so there’s IMEC, but IMEC can only expand in so far as it has commercial contracts requiring more personnel, and IMEC already has most of the available major CMOS practitioners signed up.

Crolles 2 partners NXP and Freescale, both now under the yoke of private equity firms’ tight accounting procedures, are running for cover to where they know best.

For Michel Mayer, CEO of Freescale, where he knows best is IBM, his old company. He is to give up Freescale’s involvement in Crolles2 and throw in his lot with IBM’s process technology team.

NXP is to throw in its lot with TSMC, co-founded by NXP’s former owner Philips, and a long-time partner of NXP for foundry and technology exchange.

When Jacques Chirac, President of France, opened Crolles 2 a couple of years ago, he would doubtless have been appalled if he’d had an inkling that two of the three partners in such a flagship project for European technology prowess could pull out so abruptly.

To the EC funding bodies, these defections will be a breach of trust which may imperil future funding of high-tech R&D by the European public authorities.

After the shedding of European-based labour by the big semiconductor companies following on the 2001 tech collapse, the EC authorities were already sceptical about the European high-tech companies' claims that they were long-term job creators.

Now the authorities will be even more sceptical when it comes to considering future requests for public funding of high-tech projects.

Whichever way you look at it, this is a disaster for European high-tech.

Where will the benefit of years of EU-subsidised work at Crolles go? Who will reap the rewards from the multi-hundreds of millions of Euros contributed by the EC and by European national governments?

The answer is a few rich guys at the private equity funds, and a handful of the managers at NXP and Freescale who have profits-sharing, targets-based, contracts with their new owners.

How they must be laughing at the poor old European tax-payer.

 

Article from:

http://www.electronicsweekly.com/blogs/david-manners-semi...

15:46 Gepost door Mobile blogger in Algemeen | Permalink | Commentaren (0) | Tags: crolles2, electronic, economic, imec, disaster, stmicroelectronics, nxp, tsmc |  Facebook |

07-12-06

Flanders Nano Business Alliance

Today I came across this website. It handles over Flanders NanoBusiness Alliance. The goal of this organisation is to have one voice in Flanders when discussing topics in the nano-world.

 

 

12:11 Gepost door Mobile blogger in Web | Permalink | Commentaren (0) | Tags: nano, imec, holst, flanders |  Facebook |