As a result, the RF front-end generally has become larger, more expensive and more complicated. It also typically requires more power. When the component count expands, power consumption becomes a challenge. For example, in a typical three-band design–including the transceiver, receiver and front-end module (FEM) –each supported WCDMA band requires its own amplification and filtering, and each power amplifier (PA) requires a surface acoustic wave (SAW) filter between the transceiver and PA input. Every supported WCDMA band also requires its own low noise amp (LNA) and each low noise amp needs a SAW filter between it and the RX input.

Traditional three-band design
To succeed in the next generation of wireless technology, RF hardware must become more compact, cost efficient, and capable of adapting many-band configurations. One strategy dramatically minimizes space and cost considerations: eliminating or integrating specific components such as SAW filters, LNAs (low noise amplifiers), and various matching components. This approach saves valuable board space, reduces cost, and cuts power requirements. A receiver IP can now operate in the presence of outside interferers. This kind of interference had been the main reason for incorporating SAW filters.

For example, on a triple-band WCDMA with quad-band EDGE phone, it is now possible to eliminate a significant number of components including three transceiver SAWs, three receiver SAWs, a triple-band LNA, and three baluns. Eliminating external LNAs reduces current drain, which saves power. The result is a potential saving of up to 15 percent of the RF sub-system cost due to reduced component count, and a reduction in the total board space of up to 25 percent.

Another significant advance in RF circuitry is the ability to embed RF intelligence into the transceiver. For example, this allows the baseband to start the transmitter with a single command that sets the desired channel and power level. This command sets all parameters and times the events, so that system compliance is virtually assured. The effect is to reduce the demand on RF resources. As a result, the time required from initial product development to 'first call' is shortened.

Additional savings result from reduced transactions between the transceiver and baseband, which typically use different modes, or speeds, to reduce current consumption. The key is moving between the low-speed and high-speed modes, using high speeds only when required by high data rates. Limiting RF-to-baseband communications adds significant efficiencies.

For example, the baseband no longer needs to perform sequencing or control functions. With advanced architectures applied in new RF transceivers such as the MB86L01A from Fujitsu Microelectronics, the baseband is only writing high-level commands. This means there is a minimum of RF-to-baseband communication, simplifying and speeding the transaction process. Complex transceiver control and timing calibrations are eliminated.

While single-band options remain important, future phones will require multiband implementations. Evolving this architecture will also enable use of a multi-mode PA, reducing matching components from each path, while supporting many bands and combinations.

These improvements are setting the pace for the next generation of cellular technology. The implementation cycle is now underway. Today's simpler, less costly, smaller, and less-complex RF transceiver technology is capable of supporting multiple bands, additional functions, and expanded global applications. Technological advances enable reduced component count, lower total bill of materials, and faster time to market for manufacturers–as well as phones that ensure connectivity in a global wireless environment.

About the author
Vivek Bhan has been the senior director of RF engineering for Fujitsu Microelectronics America, Inc. (FMA) since April 2009. Bhan has 11 years of experience in integrated circuit design, systems architecture and management in Motorola, Freescale Semiconductor and Fujitsu, in both the equipment and semiconductor groups. In his present position at Fujitsu Microelectronics. Bhan is responsible for managing the R&D and product development for RF and mixed signal ICs, and heads the Engineering group with more than 130 people in the areas of technology, applications, marketing, architecture, program management, RF/analog design, digital design, module development, firmware, verification, validation, product and test engineering.

Bhan received his bachelor's degree in Electronics and Communication from Punjab, India. He earned both his master's in electrical engineering and his master's in business administration from Arizona State University.He has filed several U.S. patents in the area of IC design and wireless communications. He has contributed to various magazines and publications and has received various awards. ♦