The RF challenge in a 4G multi-standard handset is the creation of a broadband network air interface that will work around the world. The hurdle here is the availability of efficient, wide-band power amplifiers to support a high performance, cost effective transmission circuit that works across all the frequency bands in use around the world.

Users are likely to want to make use of multiple functions such as GPS, mobile TV and the Internet simultaneously, so in fact a single multi-standard RF front end is highly unlikely to be the preferred architecture. Since the GPS and TV functions are receive-only, accommodating independent RF circuits for these shouldn’t be too hard to do.

Creating a broadband cellular network air interface is much harder, especially if it is to work globally. So far, ten different FDD frequency bands and four different TDD frequency bands have been defined in 3GPP that can be used for LTE (long term evolution of 3G), and it is likely that more bands will be added to this list such as 700MHz in the US.

A 4G handset will be required not only to receive but also transmit on the appropriate band anywhere in the world. Transmission is likely to pose the greatest problems, as efficient power amplifiers currently cover only one or two bands. 

Modern complex modulation schemes have the disadvantage of requiring linear power amplification, which compromises overall system efficiency because RF power amplifiers (PAs) are much less efficient when backed off from maximum power.

Conventional techniques for improving PA efficiency are inherently narrowband, and are unable to span more than a single band.
This implies five or more conventional PAs for a broadband cellular RF interface seeking to cover all ten LTE bands – adding several dollars to the bill of materials. This additional cost of multi-banding maybe be halved if this could be reduced to one or two wide band PA devices without compromising efficiency.

It is difficult to achieve flexible, high efficiency, broadband PA designs by means of RF only efficiency enhancement techniques, such as Doherty. Building Doherty solutions would require a large number of variants to cover all the different frequencies, powers and PAPR values, as these designs have inherently narrow bandwidth. 

A better alternative is a technique based on envelope tracking, that can address the efficiency issues associated with broadband ‘multi-band’ PAs – delivering an improvement in efficiency to make up for the efficiency compromises made to achieve wideband performance.

The principle of envelope tracking has been well known for many years, but it has not been commercialised until now because of the difficulty of implementing a power supply modulator capable of achieving the accuracy, bandwidth and noise specifications necessary for wideband signals such as multi carrier WCDMA, WiMAX or DVB operation. Envelope tracking technology is now being adopted in basestations, offering a dramatic improvement in efficiency, from typically low 20 per cent for a class AB amplifier to mid 40 per cent. Not the least of its attractions it the fact that it is inherently frequency agnostic, and is thus ideal for multi-band RF front ends.

The argument for envelope tracking in the handset is more about reducing the bill of materials cost than about extending battery life. Clearly, transitioning the technology from the base-station to the handset will require significant development effort.

However, I see it as a key to enabling 4G multi-standard handsets that meet user performance and functionality expectations at a price the market will accept.

Tim Haynes is CEO at Nujira