News Article

SiGe And CMOS Target GaAs Dominance Of Cellular PA Slots

Although GaAs continues to dominate in cellular PAs, several suppliers hope that their new products designed in SiGe and CMOS may soon start to appear in handsets. Tim Whitaker reports.
The power amplifier (PA) market continues to be highly lucrative for GaAs device manufacturers. GaAs accounted for about 55% of the overall market for PAs in 2003, with silicon accounting for the remainder. However, in the largest market segment, PAs for cellular handsets, GaAs enjoys almost total dominance, and accounted for well over 95% of handset PA revenues last year, according to market research firm ABI Research. Silicon LDMOS products, particularly from Hitachi (now Renesas), have achieved success in the GSM arena, but the growing prevalence of more complex modulation schemes such as EDGE and the trend towards modules are factors working in favor of GaAs. Recently, however, several companies have introduced new SiGe and CMOS PAs that could challenge GaAs in handsets.

SiGe matches GaAs performanceSiGe Semiconductor, a fabless IC supplier based in Ottawa, Canada, recently announced several SiGe-based PAs for CDMA that it hopes will challenge the incumbent GaAs technology. The company believes that its products combine the performance of GaAs with the integration, manufacturing and cost benefits of silicon. The new PAs were launched more than a year after SiGe Semiconductor announced an agreement to license IBM s cellular PA technology and hired IBM s design team. SiGe Semiconductor uses IBM as a foundry for its SiGe ICs.

The new PAs have linearity of less than -50 dBc at a peak output power of +28 dBm and a power-added efficiency of 41%. In addition to the performance attributes enabled by the use of SiGe BiCMOS, the parts are highly integrated. Each die incorporates on-chip digital or analog bias control, a power detector, a 2.8 V regulator, and all matching and harmonic tuning. As shown in the diagram, on-chip integration eliminates a number of external components including the regulator and the output detector.

Also, because the PAs vary in adjacent channel power ratio (ACPR) by less than 2 dB with 4:1 voltage standing wave ratio (VSWR) mismatch, an isolator is no longer required. "We estimate that we have eliminated around $3 from the overall bill of materials," said Jose Harrison, SiGe Semiconductor s product line manager for cellular PAs.

The new amplifiers consist of single SiGe die in a lead-frame package measuring either 4 x 4 mm for the replacement market or 3 x 3 mm for the next-generation product market. Since they contain the same die, the performance of these parts is not degraded when the package size is shrunk.

While SiGe offers high levels of integration, there are still components, such as the driver SAW filter and the output duplexer, which will remain separate from the PA module. "Our roadmap calls for us first to introduce parts with high levels of integration on the IC, and then to introduce RF modules," said Harrison. "The first step [in CDMA modules] will be to combine the PA with the SAW duplexer, so that we can match the PA output with the duplexer impedance, and operate the PA more efficiently."

Also, since neither CMOS nor silicon bipolars make good switches, SiGe Semiconductor has developed GaAs PHEMT switches that will be incorporated into front-end GSM modules that are now in development.

Ruggedness and manufacturabilityHarrison says that ruggedness has been the "skeleton in the closet" since SiGe first became commercially available. However, this problem, caused by the relatively low breakdown voltage of SiGe, seems to have been overcome. "We have spent a lot of time in developing technology to protect the PA from current and voltage excursions that occur at high mismatch conditions," said Harrison. The result is an impressive set of ruggedness claims - the SiGe PAs can survive a 10:1 VSWR mismatch at 5 V and an input power of +5 dBm with no time limitation.

Cost remains an area where SiGe proponents claim a clear advantage over GaAs (although GaAs manufacturers disagree - see RFMD sets its sights on the GSM transceiver market). The lower starting material costs and the benefits of using 8 inch wafers tend to balance out the disadvantage for SiGe of requiring a higher number of mask sets. Also, the high level of integration in SiGe ICs reduces both packaging complexity and the overall bill of materials.

What about CMOS?Since the start of 2004, two companies have announced CMOS PAs boasting very high levels of integration. Axiom Microdevices introduced a quad-band PA for GSM/GPRS, packaged in a 4 x 4 mm lead-frame module, while Silicon Laboratories announced a dual-band GSM PA in a 3.9 x 6.4 mm ceramic package. The two companies are currently embroiled in a lawsuit concerning alleged misappropriation of trade secrets.

CMOS faces a number of issues in the PA market, not least that it is ill-suited to high-linearity applications. This means that it is unlikely to be used for CDMA or EDGE. Also, for the same functionality, CMOS PAs are larger, and this tends to cancel out the advantage of having a low-cost manufacturing process. Compared with GaAs or SiGe bipolar technologies, which incorporate vertical transistors, the footprint of the FETs in CMOS increases rapidly when scaled up to accommodate high output powers.

Also, CMOS faces even greater challenges than SiGe in terms of ruggedness since it has an even lower breakdown voltage. One solution has been to develop novel circuit architectures, such as Silicon Labs version, in which voltage swings are shared across multiple transistors to preserve them from damage.

Accepting new technologiesProduct announcements are all well and good, but when will SiGe or CMOS PAs actually appear in handsets? Manufacturers have to be convinced of three key factors: viability and ruggedness of the technology; performance; and reliability of supply. Harrison believes that all these issues have been covered with SiGe Semiconductor s CDMA PAs, which are currently going through an in-house characterization phase. After this, the emphasis will shift to providing samples to potential customers for qualification.

Handset manufacturers are most likely to introduce a novel technology such as SiGe into a cost-reduction redesign of an existing handset, says Harrison. "PAs supplied as drop-in replacements, with industry-standard pin-out and packaging, reduce the risk for manufacturers and allow them to quickly bring low-cost redesigns to market," he said.

Once confidence is established in SiGe-based PAs, Harrison hopes to see rapid acceptance and growth into new handset designs. "I anticipate that we will be announcing design-ins and design wins by the end of this year, and production revenues in 2005 for the CDMA products," he said.

Ed Rerisi, director of research with ABI, says that SiGe Semiconductor is well positioned due to its links with IBM and its track record with Bluetooth and WLAN products. "If anyone is going to push SiGe into handset PA sockets it will most likely be SiGe Semiconductor," he said.

However, Rerisi is skeptical about the prospects for CMOS PAs, except perhaps in the lowest-end terminals. Both CMOS and SiGe will have to struggle to convince potential customers of the need to replace the incumbent GaAs technology. "As we have seen with GaN in infrastructure applications, it takes a long time for these new technologies to permeate the market," said Rerisi. And of course, if GaAs manufacturers can continue to improve their manufacturing efficiency and cut costs, GaAs could remain the dominant technology in cellular handset PAs for years to come.

AngelTech Live III: Join us on 12 April 2021!

AngelTech Live III will be broadcast on 12 April 2021, 10am BST, rebroadcast on 14 April (10am CTT) and 16 April (10am PST) and will feature online versions of the market-leading physical events: CS International and PIC International PLUS a brand new Silicon Semiconductor International Track!

Thanks to the great diversity of the semiconductor industry, we are always chasing new markets and developing a range of exciting technologies.

2021 is no different. Over the last few months interest in deep-UV LEDs has rocketed, due to its capability to disinfect and sanitise areas and combat Covid-19. We shall consider a roadmap for this device, along with technologies for boosting its output.

We shall also look at microLEDs, a display with many wonderful attributes, identifying processes for handling the mass transfer of tiny emitters that hold the key to commercialisation of this technology.

We shall also discuss electrification of transportation, underpinned by wide bandgap power electronics and supported by blue lasers that are ideal for processing copper.

Additional areas we will cover include the development of GaN ICs, to improve the reach of power electronics; the great strides that have been made with gallium oxide; and a look at new materials, such as cubic GaN and AlScN.

Having attracted 1500 delegates over the last 2 online summits, the 3rd event promises to be even bigger and better – with 3 interactive sessions over 1 day and will once again prove to be a key event across the semiconductor and photonic integrated circuits calendar.

So make sure you sign up today and discover the latest cutting edge developments across the compound semiconductor and integrated photonics value chain.



Search the news archive

To close this popup you can press escape or click the close icon.
Register - Step 1

You may choose to subscribe to the Compound Semiconductor Magazine, the Compound Semiconductor Newsletter, or both. You may also request additional information if required, before submitting your application.

Please subscribe me to:


You chose the industry type of "Other"

Please enter the industry that you work in:
Please enter the industry that you work in:
Live Event