ELECTRONICS COMPONENT FEATURED

Home
News Newsletter Industry News New Products Industry People Events Diary Products Buyer's Guide Articles Tech Articles Archives Recruitment Jobs (New) This Site Advertise Contact Us Terms & Conditions Disclaimer Google Intranet login UserName: Password:	Featured Article - Efficient RF design in a digital world By Dave Deighton – Senior Director Customer Engineering, Nujira Limited Publication date: 14 April 2009 Featured Article - Efficient RF design in a digital world By Dave Deighton – Senior Director Customer Engineering, Nujira Limited New standards of digital information transmission across RF signals using OFDM modulation in standards such as WiMAX, LTE, DVB and DAB offer high spectral efficiency through use the amplitude of the transmitted signal to encode information. In consequence, they exhibit a much higher Peak to Average Power Ratio (PAPR), so that the RF amplifiers spend less time in ‘efficient’ saturated mode, and the energy they use in transmission increases dramatically. These new standards are often shoe-horned into overcrowded RF bands around the world, and as a result terminals need a multi-band RF front end if they are intended for global deployment. RF amplifiers can end up consuming as much as half the power in a high speed modem. Since there are few efficient, wide-band power amplifiers, designers end up implementing five or more PAs to provide a multiband design that will cover all the anticipated requirements. There are a number of techniques available that can partially address these challenges, such as DPD, linearization and Doherty. However, none of them compensates fully for the efficiency lost due to the higher PAPR, they carry a relatively high design overhead, and some, like Doherty are inherently narrowband, and are unable to span more than a single band. It is time to take a fresh look at the design of the RF transmission amplifier chain. Cross-industry challenge The power of the transmitted signal increases by an order of magnitude from the cellular handset to the cellular base station to the broadcast transmitter – but the challenge and the base of available potential solutions is essentially identical. Though the highest volume applications for RF transmission today are cellular handsets, cellular networks and digital broadcast, the arguments and the solutions described here are applicable to all digital transmissions, including medical, military, industrial and other environments. RF power amplifiers are typically classic Class AB configurations, and accordingly are only at their most efficient when the RF envelope waveform is close to peak power. This is not usually a problem with traditional signals like 2G GSM, where the amplitude is constant, so the PA operates in this high efficiency mode all the time. However, when amplifying amplitude modulated RF signals with high crest factor, it is less efficient, typically in the range 15-25% for W-CDMA and WiMAX. As can be seen, for much of the time the signal power lies well below the peak power and hence the device is operating at low efficiency. Possible solutions for improving PA efficiency A number of techniques are now being used to improve PA efficiency in through the RF amplifier chain including: • Crest Factor Reduction (CFR) and Linearization (digital pre-distortion, DPD) • Doherty • High Accuracy Tracking of the Modulation Envelope Crest factor reduction and linearization Crest factor reduction can make a useful contribution to efficiency by suppressing peak signals, thus allowing the amplifier to operate nearer its peak power and maximum efficiency. Linearization techniques, such as digital pre-distortion build on this, by compensating for known nonlinearities in the RF output stage. This allows the amplifier to be driven further into compression, providing better efficiency whilst still meeting adjacent channel and error vector magnitude (EVM) requirements. Optimum improvements are obtained by incorporating pre-distortion in a system that embodies sampling of the output signal within a feedback loop. In this way the system can compensate for changes in amplifier characteristics adaptively with temperature, power supply and time. Crest factor reduction and DPD rely on intensive signal processing techniques, which need to be implemented in the baseband. Running a 2W DSP to double the efficiency of a 40W transmitter delivers a worthwhile benefit. For a handheld transmitter like a cellular handset, this might translate to a 500mW DSP linearizing the power of a transmitter running at 250mW full power which gives a limited advantage at best, and might easily reduce efficiency when the handset is close to a base station and transmitting at minimum power. Doherty configuration For the past three years the Doherty amplifier architecture has become the “amplifier of choice” for new wireless transmitters, because of its ability to enhance efficiency in complex modulation schemes. The classic Doherty configuration comprises two amplifying devices driven in parallel, with their outputs combined. The main or ‘carrier’ amplifier is typically a standard Class AB device and provides all the output power until it enters it nonlinear region. At this point, the second ‘peaking’ amplifier (normally operating as Class C) is switched on to provide additional power. The outputs of the two amplifier sections are combined using an impedance inverter, providing the novel feature that the carrier amplifier can continue delivering power into the load as the signal output increases. Careful design is required to cancel out the two amplifiers’ nonlinear transfer characteristics, creating a near-linear characteristic when the outputs are combined. Doherty designers also need to be mindful of coordinating the changeover in the breakpoint region, by adjusting the peaking amplifier’s start point and gain expansion so that full power is achieved at the same point as the carrier device. The efficiency gains for Doherty PAs come at a significant cost. Designers need to address the difficulties in maintaining matching and linearity with temperature variations and drift over time. Most crucially, the phase sensitive combining network between carrier and peaking amplifiers makes Doherty amplifiers essentially a narrow-band solution covering for example just one WCDMA band such as 1920-1980MHz. So, building Doherty solutions would require a large number of variants to cover all the different frequencies, powers and PAPR values. At best, a single PA using these techniques can cover one or two of the frequency bands required in a 3G-plus base station or mobile phone. With handsets in particular bill-of-materials is even more crucial than power consumption and Doherty simply cannot rise to the challenge. High Accuracy Tracking High Accuracy Tracking (HAT™) uses a completely different technique to enhance efficiency – and it does so over a wide frequency range and across multiple modulation modes. Instead of optimising a final RF stage power transistor supplied by constant voltage, the supply is changed dynamically, modulated in synchronisation with the envelope of the transmitted RF signal. Doing this ensures that the output device stays in saturation – its most efficient operating region for a large portion of time. With a fixed supply voltage, there is a great deal of excess power over the required output waveform and this is dissipated as heat. Tracking the RF signal envelope closely with the supply voltage makes for a much lower differential, dramatically lowering the excess power. Supply voltage is varied by a power modulator device, which replaces the normal DC-DC converter delivering power to the output transistor. HATTM is based on envelope tracking, a principle that has been known for half a century, but has not previously been implemented in a technically or commercially viable solution. Critical performance issues include tracking accuracy, modulator efficiency, stability, compliance with spurious-signal and noise specifications, and bandwidth for multi-carrier support. By developing a groundbreaking High Accuracy Tracking™ technology, Nujira is the first vendor to successfully meet these challenges. HATTM based solutions offer an impressive improvement in efficiency, going from typically a low 20 percent for standard a class AB amplifier to mid 40percent with HAT. Multiple base station and Digital TV transmitter manufacturers are at an advanced stage of adopting this new technology in their products and Nujira is now developing a solution for handsets. A typical base station system implementation involves the addition of the HATTM modulator module, a 70mm X 70mm X 18mm metal housing, plus the addition of a digital connector allowing the envelope signal to be routed from the digital baseband and passed to the HAT module. In addition, some minor redesign of the PA layout is needed to ensure optimal matching and hence efficiency. Other design considerations that have to be dealt with compared with a fixed-supply-voltage solution are the need for a high-bandwidth drain feed to cope with the modulated supply plus the need to adequately minimise RF leakage from the drain. To retain compliance with demanding noise and spurious specification, the power modulator tracks the RF signal envelope with utmost accuracy in both timing and amplitude. It does so by calculating the amplitude from the digital signal (√(I² + Q²)) and applying a simple function to arrive at the optimum instantaneous drain voltage. In parallel, a delay is calculated and applied to the RF signal before it is input to the amplifier, cancelling out the delay in the modulator. Optimising average power output There is another important consideration with hand held terminals, compared with essentially static applications such as base stations or DVB broadcasting transmitters. Because the mobile device is roaming, it needs to optimise power consumption by varying the average RF output according to its distance from the base station. In the same way that HATTM optimises efficiency across a probability distribution fixed by the transmission mode (modulation type); it also provides maximum efficiency throughout a probability distribution set by average output power requirements across a mobile network cell. The result, again, is a huge improvement in the PA’s power dissipation for a given output power. In fact, calculations show that the HAT-enabled PA uses half the battery power of those using conventional MMICs. Conclusion To put the above discussion into context, NTT’s published CO2 emissions in 2006 were about 3.8 million tonnes – or around 3% of Japan’s total. Industry estimates suggest that 86% of this figure would come from powering its transmission network, and we estimate about half of that from the transmission circuit itself, with the associated cooling systems making an additional contribution. Though DPD, linearization and Doherty can all make an impact on this figure, only HATTM is capable of fully compensating for the inherent inefficiencies of transmitting OFDM signals, and reversing the trend of rising RF transmitter energy use. For Further Information, Please Visit http://www.nujira.com Send to a Colleague! Your Email: Their Email: Comments: Please enter your email address!" controltovalidate="txtemail" isvalid="true"> Please enter your email address! Please enter a Valid Email Address!" controltovalidate="txtemail" validationexpression="\w+([-+.]\w+)*@\w+([-.]\w+)*\.\w+([-.]\w+)*" isvalid="true"> Please enter a Valid Email Address! Please enter your colleague/friend's email address!" controltovalidate="txtemail2" isvalid="true"> Please enter your colleague/friend's email address!
© Electronics Components World - email: info@electronicscomponentsworld.com  Electronics Production World