Manufacturing Industry

Mixed-signal processor technology solves common digital problems

Electronic News, August 18, 1997 by Bellisario, Afarin

The goal of successive generations of electronic products is for each new member to be smaller and cheaper, and offer more functionality and features. It is widely accepted that the integration of semiconductor components will help achieve these goals. In systems containing both analog and digital components, the ultimate integration involves mixed-signal integration. Whether for data communications exploding around the Internet, wireless communications, or consumer electronics, the trend toward analog and digital signal processing integration is clear.

Mixed-signal processing can help designers realize the marketing requirements of the end product by integrating analog ICs, mixed-signal codecs and digital signal processors onto a common silicon device known as mixed-signal processor (MSP). When ADI introduced its first mixed-signal processor in the early 1990s, products that interfaced through the human voice were the most efficient application of MSP technology. In that time, ICs designed to process signals within voiceband frequencies were most likely to hit the integration sweetspot enabling smaller, faster, cheaper end-systems.

Today, however, having invested in MSP process technology and IC design expertise, ADI is able to offer MSP which offer size, cost, speed and power consumption advantages in a wider range of applications.

Not a Simple Marriage

Achieving a high-performance and cost-effective mixed-signal part is not a simple marriage of analog and digital components. The list of technology hurdles includes the difficulty of achieving a high-performance analog functionality using a CMOS technology optimized for digital processing, the "digital" noise floor generated by the crosstalk between analog and digital parts, and designing efficient test methodology.

Moreover, analog IC design methodology traditionally had little in common with digital IC design. Therefore, successful and cost-effective integrating of high-speed digital processors and analog converters requires expertise in the science of manufacturing process technologies, the art of efficient and elegant chip design, and a thorough knowledge of systems architecture. The partitioning of the system is crucial to achieve cost efficiency and performance.

The value of reducing the number of chips through mixed-signal processing is reduction of cost in ownership, improved product design, performance and reliability. Decreasing the number of chips per product reduces manufacturing complexity which increases the manufacturing throughput while reducing both manufacturing and procurement costs. Smaller footprint and lower parts count reduce the physical size of the product and increase the end product's reliability by reducing the likelihood of part failure. Additionally, the use of MSPs can reduce the time-to-market significantly by simplifying development board design and eliminating the need for designing complicated analog-to-digital interfaces.

Of particular significance to portable or battery-powered products is the power consumption. Mixed-signal processors consume considerably less power.These advantages make MSPs well suited to cellular telephones, pagers, digital tape records, voice mail systems, digital answering machines, and cordless telephones. The addition of sophisticated power management circuitry enhances the power conservation.

Fixed or Programmable

The digital processing element of an MSP can be fixed function (such as filtering, FFT, etc.) or programmable, (i.e., a general purpose DSP). The fixed-function MSPs, also known as mixed-signal ASICs, are common. IC vendors such as SGS Thomson, Motorola, and Crystal Semiconductor all offer standard ASICs performing standard audio band processing. However, only a few companies, including Analog Devices, offer programmable mixed-signal platforms.

The programmable MSPs provide product differentiation capability and flexibility to OEMs, while preserving the size and cost advantage of a highly integrated IC. Similar to DSPs, ROM coded programmable MSP provides a quick way to bring highly integrated application specific standard ICs to the market. A single programmable MSP can provide many functions, such as voice processing and modems. This will increase the "space efficiency" of the product, and allow the OEM to provide multiple functionality in a small package. Given the advantages of programmable MSPs, it is only a matter of time for these products to become platform of choice for design, development, and manufacturing of most electronic products.

Mixed-signal processing is an excellent solution for portable or space-constrained designs. By combining multiple chips into a single MSP, companies can reduce time-to-market and create competitive advantage. When choosing a supplier of MSPs, it is paramount to ensure that the vendor not only have a history of success in both analog and digital components design and fabrication, but also have successfully designed and manufactured mixed-signal processors for a wide variety of applications. To assure that the training and code investments are preserved, OEMs should also make sure that the programmable MSP of their choice is supported by a full suite of easy to use tools, preferably the same ones designed to support the DSP core.


 

BNET TalkbackShare your ideas and expertise on this topic

Please add your comment:

  1. You are currently: a Guest |
  2.  

Basic HTML tags that work in comments are: bold (<b></b>), italic (<i></i>), underline (<u></u>), and hyperlink (<a href></a)

advertisement
advertisement
  • Click Here
  • Click Here
  • Click Here
advertisement

Content provided in partnership with Thompson Gale