Designed in the late 1980s, digital audio broadcasting (DAB) technology is now coming up to potentially replace traditional analog forms of broadcast radio. Taiwan-headquartered IC design startup Future Waves proclaims one of its missions as capitalizing its management expertise in the semiconductor industry and in the mobile broadcast industry in order to timely develop next generation products for the digital broadcasting market, and DAB is a priority for the company. Several months ago, the company launched its first product, a single-chip receiver for DAB. DigiTimes.com recently an opportunity to talk to Glenn Vandevoorde, CEO of Future Waves. Last year, he stated that the company will develop a range of products for different digital broadcasting standards including DVB-H (Digital Video Broadcasting – Handhelds), but our talk this time was mainly dedicated to DAB.
Q: Your company is very young, and it represents a new wave of Taiwan IC design houses. How was your company born? Who are your strategic investors?
A: Future Waves was born as a joint venture between UK-based Toumaz Technology, a spin-off company from Imperial College London, and Applied Bionics of Singapore. Toumaz Technology is a semiconductor design company focusing on development of technology platforms. Applied Bionics is a consortium of industrialists targeting commercialization of IP in the Far East. Through one of Toumaz’ developments (RF CMOS chips for digital tuners), Applied Bionics identified a commercial opportunity in this area. Future Waves was then set-up to turn the technology platform into a range of products. Future Waves has an exclusive and worldwide license to Toumaz's IP platform. So even though the company can be considered young in terms of operations, the technologies it employs have a long development history.
Q: In creating your own business model, have you considered the experience of any older IC design houses? Did you learn anything from their development paths? Is there any IC design house in the world that you see as a good example for Future Waves to follow?
A: Future Waves’ expertise and differentiation from our competitors lies in RF CMOS design. Good examples of similar design houses (during their earlier stages), which have become successful through focusing on RF CMOS developments, would be SiLabs (GSM and FM solutions), Cambridge Silicon Radio (CSR – Bluetooth) and Atheros (WLAN).
Q: Do you plan to go public in the future? Do you plan to participate in any industry associations?
A: IPO plans would be for around the end of 2008 or beginning of 2009. We are an active member of WorldDAB and AsiaDAB, two organizations that aggressively promote the acceptance of DAB and DMB (Digital Multimedia Broadcasting) standards worldwide.
Q: How many employees do you currently have? How long is your design cycle?
A: As a fabless design house, we currently have 29 employees in Taiwan, a support team in Korea and have recently set-up a UK-based R&D team for further developments of our RF CMOS platform. Our design cycle is typically 9-12 months.
Q: So far, your company has introduced one product, the Fenix FNX14701 RF tuner chip for Digital Audio Broadcasting (DAB), and samples of the product have supposedly been available already for several months. Have you started volume production? What is your sales expectation for the near-term future? Do you have any revenue target for 2006?
A: Engineering samples are available now and production samples will be available at the end of this quarter. Functional samples have been available since October. Our sales expectation for 2006 for the DAB and T-DMB (Korea) markets is 1.5 million units.
[Editor’s note: Korea launched its T-DMB (Terrestrial – Digital Multimedia Broadcasting) service in December 2005. T-DMB is a standard of the European Telecommunications Standards Institute (ETSI). It is based on the DAB standard known as Eureka 147, which was developed as a research project for the European Union.]
Q: How would you describe your current relationships with Taiwan hardware makers? Do you prioritize the Taiwan community compared to companies from other regions?
A: We are based in Taiwan to be close to customers. We do prioritize Taiwan customers because we can provide very good support locally. There are many potential customers focusing on module solutions, such as Gyrosignal, Orthoradio and MTECH, and we also see opportunities to cooperate with system integrators.
Q: The Fenix FNX14701 chip is manufactured under a 0.18-micron process. What company is your manufacturing partner?
A: We manufacture at Taiwan Semiconductor Manufacturing Company (TSMC).
Q: What company is your partner for packaging chips?
A: We use Siliconware Precision Industries Ltd (SPIL) and Advanced Semiconductor Engineering (ASE) for packaging.
Q: Do you consider using any other packages than the current Quad Flat No-lead (QFN)?
A: Other packages can be made available upon requests from customers, but thus far the QFN package seems to be the most in demand.
Q: Last year, you stated that Future Waves is targeting digital tuners for portable applications with its RF platform. Why does the FEK14701 evaluation kit support only Windows operating systems, which are more focused on desktop PCs?
A: The evaluation kit is only there to evaluate our solution and provide customers with an easy platform to understand the application. The end application (product) does not need to be Windows-based to function.
Q: According to the specification, the Fenix FNX14701 chip features itself as a single-chip RF solution for receiving DAB L-band, DAB band III and FM signals, and the two mentioned DAB bands are reserved for European systems compliant with the Eureka 147 standard. Should this be understood as your exclusive focus on the European DAB developments or do you also plan to support US digital radio schemes, including S-band signals and the In-Band On-Channel (IBOC) method of transmitting digital radio and analog radio broadcast signals simultaneously on the same frequency?
A: We focus on DAB solutions now, but also support L-band functionality for digital satellite radio networks, such as WorldSpace. Our roadmap includes products supporting IBOC and S-band radio in the US, but this is not yet provided by our FNX14701 chip.
Q: The specification also mentions that the FNX14701 outputs real intermediate frequency (IF) signals at 2.048MHz (DAB) and 14.336MHz (FM). Can you explain this? Why is this important?
A: This is to be compatible with all existing baseband solutions. Even though we have a zero IF architecture and provide in-phase and quadrature (I/Q) outputs, we upconvert again to 2MHz to be compatible with existing baseband solutions that do not support I/Q inputs.
[Editor’s note: a zero IF (or direct conversion) architecture implies a one-stage conversion from radio frequency to zero IF, the RF signal being mixed with in-phase and quadrature components of a local oscillator working at channel center frequency. This approach is different from the double superheterodyne architectures with two-stage IF filtering, where the intermediate frequency is usually much higher compared to the signal bandwidth and the first IF filter stage needs to have enough selectivity to provide adequate rejection of the image frequency of the second IF conversion stage. Utilizing a zero IF architecture reduces filter stages that makes it possible to use less off-chip components, such as IF surface acoustic wave (SAW) filters. It also reduces power consumption, makes possible the integration of channel select filters on-chip, reduces a solution's footprint, and cuts down the bill of materials and manufacturing costs.]
Q: Future Waves licenses the RF CMOS technology platform and IP developed by Toumaz. Does your company have its own R&D? Can you reveal your R&D budget?
A: Yes, we have R&D in Taiwan and have just set-up an R&D center in the UK. Our R&D budget is around US$5 million per year.
Q: Does the FNX14701 incorporate any special method of decreasing power consumption? Is Future Waves working to make further progress in this direction?
A: We use RF CMOS for more integration into silicon. We also use direct conversion techniques, which reduces power consumption. Our RF receiver is also designed for very fast switch on and switch off to enable maximum benefits for power cycling.
Q: Besides the currently supported I2C, will your solutions support more I/O buses?
A: Yes, our next-generation chipsets will provide more I/O interfaces.
Q: What kind of opportunities do you see for Future Waves in China? Many industry observers agree that China's IC design community has very good prospects. What do you think about this? Do you consider design houses there as competitors, or maybe some of them are potential partners?
A: We have several strategic customers in China and are actively involved in promoting DAB in China through WorldDAB and AsiaDAB. The IC design skills in China have been progressing very rapidly, but we do see that in terms of RF CMOS developments, and RF developments in general for that matter, skills are relatively hard to find. This has been a challenge in Taiwan as well. Attracting local talent has not been easy, especially as a start-up. We have and are actively tackling this issue by offering an attractive working environment, both in terms of remuneration and stock options and continued training through our Taiwan and UK R&D capabilities.
CEO of Future Waves Glenn Vandevoorde: “We focus on DAB solutions now, but also support L-band functionality for digital satellite radio networks, such as WorldSpace.”
Photo: Company.
Article edited by Stephen Taylor
