ChipCenter Questlink
SEARCH CHIPCENTER
Search Type:
Search for:




Knowledge Centers
Product Reviews
Data Sheets
Guides & Experts
News
International
Ask Us
Circuit Cellar Online
App Notes
NetSeminars
Careers
Resources
FAQ
EE Times Network
Electronics Group Sites

  ASIC News

    Tech Note

ASIC Main | Archives

Page 1 of 3

The Wireless Dilemma—Adding Features Without Sacrificing Cost, Standby, and Talk Time

by Adelante Technologies,
Abdojstraat 34
3001 Leuven, Belgium

+32 16 39 14 11
www.adelantetech.com

Accelerating Digital Processing Performance with Ultra Low Power and Minimal Die Size

Jump to...
Galaxic™ DSP Technology
   Reconfigurable, Extendable DSP Cores
   Lunar™ Subsystem Library
   Atmosphere™ Development Environment
Reconfigurable, Extendable Saturn DSP
  Core™ for Wireless Handsets
   Architecture
   Compact 16-bit Instruction Set
Reconfigurable Instructions and
  Extendable Architecture
   Application-Specific Instructions
   Application-Specific Execution Units
   Application-Specific Coprocessors
Conclusion

As next-generation mobile phones evolve from supporting only voice transmission to the transmission of audio, images, and video, substantial baseband processing power will be required to handle the increased digital transmission, reception, compression and decompression of substantially larger data streams. General Packet Radio Service (GPRS) and High-Speed Circuit Switched Data (HSCSD) use multiple time slots, which are processed in parallel to increase the data rate. They increase baseband processing alone by a factor of eight, and they are becoming standard features in today's mobile phones.

If handsets were neither handheld nor cost-sensitive, adding these features would be a matter of either turning up the processing clock or using a more powerful processor in the phone. Sadly, these solutions are unacceptable in the handset market because they will increase power consumption, product size, and/or product cost. Product cost and battery life are the two most important attributes of a mobile handset. Wireless service providers commonly give handsets to end-users for free in order to get customers. Since the providers give the phones away, they want the cheapest implementation consistent with acceptable performance. After cost, the most important features are standby time, talk time, and the phone's feature set, in that order. The highest prices are commanded by the phones with the longest battery life and the smallest overall size.

The question is "How does a designer meet increasing processing requirements for next-generation phones without sending power consumption, cost, or end-product size through the roof?"

Traditionally, performance increases have been achieved in one of three ways:

  1. increasing the clock frequency;

  2. adding extra processors; or

  3. selecting a more powerful (and more expensive) highly parallel, VLIW processor with a memory-hungry VLIW instruction set.

Increasing the clock frequency can require a higher supply voltage that increases the power consumption exponentially. VLIW processors enable increased performance at lower clock speeds, but they are expensive from a total system point of view since they cost more and require substantially more program memory to store the larger instructions. Larger memory results in higher cost and greater power consumption, as well as a larger footprint design. Adding a second processor (e.g., for speech processing) again increases the cost, size, and power drain of the end-product. To wit, all of these traditional solutions end up destroying the most important competitive advantages of the phone—low cost and long standby and talk times.

This inevitable trade-off between power and performance suggests that the traditional approach of using off-the-shelf DSP processors and cores may not be suitable when performance and cost and power consumption are equally important. Even the lowest power versions of standard DSPs consume a lot of power when the clock frequency is high enough. Power consumption of only 0.5 mW/MHz translates into 200 mW with a 200 MHz clock. A handful of standard DSPs that are fabricated in extremely small process technologies can operate with a fast clock and low power consumption. However, the processes and the devices are extremely expensive, violating the low-cost rule for handsets.

Adelante Technologies, the result of a merger between Philips-DSP Division and Frontier Design, a leading DSP IP and EDA vendor, is currently developing DSP solutions that solve these challenges. Adelante's Galaxic™ DSP Technology provides digital signal processing solutions that allow performance to be radically accelerated while maintaining ultra high performance and minimal silicon area.

Adelante Technologies' Galaxic™ DSP Technology
Back to top

Adelante Technologies' Galaxic™ DSP Technology is a comprehensive offering of digital signal processing solutions for high-performance, low-power, and cost-sensitive DSP applications. It includes several families of existing and planned DSP cores, associated subsystems, development environments, EDA tools, and design services.

Reconfigurable, Extendable DSP Cores
Back to top

Galaxic DSP Technology cores are designed for high-performance, low-power, and low-cost digital signal processing applications in the areas of wireless, digital control, infotainment, and networking. The Saturn DSP Core™ is the first Galaxy DSP family, and is targeted to cost-sensitive, high-volume applications such as wireless handsets and controllers for DVDs, CDs, and hard-disk drives. During the next 18 months, Adelante will introduce additional DSP core families that are optimized for other high-performance, low-power markets.

Lunar™ Subsystem Library
Back to top

Integrating a "naked" DSP core into the end-product SoC with external memory, peripherals, I/O, and multiple processors, and then verifying the whole embedded SoC design can be a daunting task that requires up to a year of engineering. All Galaxic DSP cores are pre-integrated with a complete, configurable DSP subsystem that includes on-chip JTAG debug capability, a DMA controller, arbiters, paging, built-in self-test (BIST), monitor, program and data memory, and program and data cache. Interfaces to external memory, peripherals, and processors (including the AMBA bus) are built-in and fully verified.

Atmosphere™ Development Environment
Back to top

The Galaxic DSP Technology also includes the Atmosphere™ Development Environment, a comprehensive suite of code development and debugging tools, as well as EDA support for the design of application-specific extensions to both the instruction set and the core's internal hardware. The Atmosphere Development Environment includes all DSP code development tools, including compilers, linkers, debuggers, and profilers. An instruction set simulator (ISS) facilitates early code debugging by providing single-step program execution, the setting of breakpoints, and access to all memory and registers in the core. The debugger also works with the on-chip JTAG run-time debug facilities available on the Lunar subsystem, and allows single-step code execution in the processor itself.

Atmosphere interfaces to development tools from Mentor Graphics (XRAY Debugger, Seamless Coverification Environment, Modelsim HDL simulator) and AXYS (fast behavioral simulation models, MaxSim™ multi-core C/C++ simulation environment).

Next >>

Click here to get your listing up.

Copyright © 2003 ChipCenter-QuestLink
About ChipCenter-Questlink  Contact Us  Privacy Statement   Advertising Information  FAQ