blank
blank
only search Cheshire Henbury

Cheshire Henbury's website is structured around several sub-sites to accommodate the large amount of content. Please pick a topic of interest from the above menu and begin to explore and learn. Or use the Google Search box to the left.

Paul T Kidd's Embedded Microprocessors Pages

Main Home > Legacy Content Home > Embedded Microprocessors Home > Article on Microelectronics

Article on Microelectronics

Article on Microelectronics

 

Introduction

Microelectronics is one of the fundamental enabling technologies driving the development of the Information Society, but what are the problems resulting from physical, technological and economical limits?

The main issues are how to address complexity, how to address design re-usability, modular verification and testing, skilled resources and a vision of tomorrows applications and the need to improve design in order to be more profitable.

Markets

Technology inflections are creating markets. Early bipolar transistors enabled the mainframe computer business. In the 1980s microprocessors and DRAM memories enabled the PC business. The current technology is system on a chip. This is about putting everything on one chip and this will lead to new markets - for such things as mobile and handheld devices, set-top boxes, digital TV, and so on.

In the PC era the key issue was how to increase the performance of silicon in terms of speed and number of transistors on an integrated circuit. Now the challenge is to decrease the total cost system. The old PC era was about a final end markets of around 100 million units, but the new system on a chip era is about serving markets characterised by billions of units. These large numbers are accounted for by the replacement of existing analogue devices (such as TVs) as well as the introduction of new devices, for example e-books.

The semiconductor industry now drives the wealth of nations. Europe is strong in certain areas, which are relevant to the new era (digital TV, mobile phones). The latest figures suggest that the European semiconductor business is positioned for growth in many areas - telecommunications, automotive, consumer as well as industrial - and many firms are already leaders in these areas.

 Technical and Human Resource Challenges

The future technical challenges over the next 20 years lies in the area of cost. Traditional cost reduction has focused on the shrinkage of the size of the transistor and increasing the number of chips per wafer. There is a need to increase the wafer size, which is a technical problem. There is also a need to continue to reduce the size of the transistor. This requires financial resources.

After 10 to 20 years a problem with physics arises as the limits of silicon are reached. By 2010 the size of a transistor will have reached 20 nano-metres and at this size quantum effects start to manifest themselves. At that point new technologies will be needed - nanotechnologies.

In about 10 years there will probably be a mixture of micro and nano. Increases in the density of integrated circuits may still be pursued, or a bottom up approach may be adopted involving use of nano materials mixed with microelectronics.

A major challenge that has to be faced is human resources - ensuring that industry has the necessary supply of skilled people to support the expected growth and to solve the technical problems.

Institutions where research, education and training can be combined are one possible means of dealing with both the technical and human resource challenges.

Integrated Circuit Design

The challenge of integrated circuit design is mastering complexity. Another issue is how to test integrated circuits containing one hundred million transistors per square centimetre, given that such a test should only take 1 second!

In the 1990s the focus was on speed (increasing the clock frequency) and density (the number of transistors on an integrated circuit). In the first decade of the 21st century the focus is shifting to power consumption efficiency (because of the growth in handheld devices) and the handling complexity (reducing the design and manufacturing costs). Technologies need to be developed to help people in the design field.

Currently the industry is facing a design crisis. Process technologies have been offering efficiency increases of the order of 59%/year. However, efficiency increases in the design area are only a low 25% per year. There is a design gap that prevents design using the technologies to full effect in the time available. Therefore there is a need for a new design paradigm to close this gap.

Design reuse provides a means of closing this gap. Standard cells are now well established and reuse of IP blocks is now being used. New ideas concerning architecture reuse are beginning to emerge and some research is starting on integrated circuit reuse.

There are a number of challenges that must be dealt with. First is how to master the complexity of communication, with high-speed massive data transport, making sure all the data is available at the right time in the right place. Concepts from telecommunications, such as local communications and global communications, are entering the field of IC design. Second is the need to verify the design. Also at the electrical level issues like noise have to be dealt with.

 Applications Design

In the field of applications in the automotive area, a number of design challenges are posed. The importance of automotive electronics is increasing and electronics is being used in several areas including powertrain, dashboard, navigation, safety, and in-car entertainment.

In automotive applications, as in other areas of electronics, economies of scale are important in order to pave the way for lowers costs and higher quality. European firms are at the forefront of innovation in the field or automotive electronics. The value of electronics within cars is increasing significantly which represents a significant opportunity for growth and profitability.

One of the fastest growing areas is that of in-car entertainment systems. Traditionally this has means a radio, with either an integrated cassette or CD players. Stand-alone navigational systems are also becoming a significant item, and growth in this area will be important in the coming years. However, the systems of the longer-term future will be integrated multimedia systems incorporating several different functions - entertainment, communications, navigation, diagnostics, etc. Services to support some of these functions will also be introduced - e.g. location based services advising about the nearest doctor, hospital, 4 star hotel etc.

A key problem is creating the volume per application and per Application Specific Integrated Circuit. Customers, that is the car firms, have a tradition in this area of expecting and demanding unique products. In order to achieve economies of scale and reduce costs this sort of thinking must change (a culture change issue). Standardisation is needed, with good platforms providing lots of re-use potential covering all the multimedia applications in vehicles, with only a small amount of customer specific features. Modular designs, design rules and standard top-level architectures are needed. Systems should be open, and be capable of being scaled and updated in the future. By placing some functionality in software it will also be possible to carry out upgrades to existing products, which may also lead to an additional source of revenues.

Partnerships with other firms with the necessary competencies are needed in order to achieve this goal of design re-use. Understanding the balance between centralised systems and distributed systems is also important in order to understand costs factors.

Conclusions and Future Directions

The costs of researching, designing and fabricating silicon devices are showing signs of increasing. Therefore research is needed into reducing the cost of design through such techniques as reuse of design, modular verification, and other aspects that can help to keep costs under control. There is a need for a more multidisciplinary approach in the semiconductor industry and a need to improve design methods to reduce costs and improve productivity.

Skill shortages are a key factor in preventing the realisation of the vision of ubiquitous computing and networking. A shortage of skilled technologists and engineers will slow down technology development. More investment needs to be made in education and training.

 

 

Some of Paul T Kidd's Books

Book Covers

Legal Notice: The information posted on the web site is designed to provide accurate and authoritative information in regard to the subject matter covered. Every effort has been made to ensure the accuracy of the information contained in the web site. The information is believed to be correct at the time of publication. Cheshire Henbury cannot however accept any responsibility for the completeness, accuracy and relevance of the information. Information is published with the understanding that publication does not represent the rendering of advice, consulting or other professional services. Specific application in a particular organisation is the sole responsibility of the representatives of that organisation. If expert advice is needed, the services of a competent person should be sought. Please read our terms and conditions (opens in a new window) for use of this web site.

Cheshire Henbury

Address and Phone Details (opens in new window)

Email: Contact form (opens in a new window)

Web address: www.cheshirehenbury.com