Envelope Tracking Mitigates LTE RF Front-End Power Issues

Over the last decade, 2G/3G RF front-end designs have evolved to a mature and relatively efficient state, with little innovation over the last few years. Instead, with the rollout of LTE networks on the horizon, the race was on for chip vendors to develop their LTE baseband modem and transceiver chipsets. Today most of the chipset challenges have been addressed, with all the major vendors releasing viable products for the first wave of 4G handsets onto the market.

However, the relatively slow pace of RF front-end development has meant that handset manufacturers and operators still face the significant RF system challenges presented by LTE. These challenges are having a major impact on the consumer experience of LTE phones and networks. Facing problems such as poor battery life, antenna performance, and thermal issues, the industry is now putting RF front-end development at the top of its agenda to enable first-class 4G handsets.

Download this article in .PDF format
This file type includes high resolution graphics and schematics.

RF Front-End Challenges

LTE networks and signals present a number of problems for the RF front end in handsets as the signal characteristics differ so extensively from 2G and 3G standards. Modern digital modulation techniques are compressing more data bits into every RF channel, resulting in more complex waveforms with higher “crest factors,” commonly expressed as peak to average power ratio (PAPR).

LTE signals have a very high crest factor (7.5- to 8-dB PAPR), resulting in a much higher peak power requirement for the transmitter. This increased requirement necessitates a larger power amplifier (PA) and increased supply voltage, leading to a corresponding increase in wasted energy. With the RF PA in 4G handsets becoming such a significant consumer of power, usage times for an LTE device have dwindled to mere hours.

Adding to the signal challenges, handsets also need to simultaneously support different systems (TD-LTE, FD-LTE, WCDMA, HSPA, GSM, etc.) whilst supporting an expanding number of frequency bands (LTE is currently deployed in 20 different bands globally), leaving engineers with an extremely complex RF system on their hands.

The Frequency Band Dilemma

Operators are demanding truly global roaming 4G handsets that cover 14 to 15 of the more than 40 frequency bands allocated in the highly fragmented LTE spectrum, alongside the 3G and 2G bands. This isn’t currently possible.

The traditional discrete narrowband PA method is no longer feasible, as dedicating a PA and RF filter to each band inflicts severe size and power penalties on handsets. So, up to nine variants of some handset models are created to provide for all regions and operators.

This is a particular problem in the U.S., where LTE spectrum is very fragmented. For example, Apple had to implement dedicated variants of the iPhone 5 for both Verizon and AT&T to ensure the required bands were covered. However, this is a logistical challenge for manufacturers, as it complicates the design, validation, production, and inventory management processes.

One obvious solution is to implement several broadband PAs covering far more frequency bands in a single handset. Unfortunately, this leads to further performance degradation, due to the low power efficiencies achieved by today’s broadband PAs, making them unattractive for commercial deployment. There is a 5% to 10% decrease in energy efficiency between a narrowband and broadband PA, which is too high a power handicap for phones where power is critical.

Mobile handset engineers are in a no-win situation, as both narrowband and broadband approaches are inadequate to meet the heavy demands being placed on LTE handsets. Consequently, designers have been forced to assign more space and power to the RF front end, affecting the cost, size, and battery life of handsets.

Is There A Solution?

Ideally, handset manufacturers want to implement a single multimode/multiband PA solution that surmounts the space and power problems caused by traditional PA approaches. Before this can happen, new techniques have to be designed to improve the power capabilities and efficiency of PAs.

The most promising technique among the new wave of RF innovations is envelope tracking (ET). This power supply technique replaces fixed dc supply voltage to the RF PA with a very high-bandwidth dynamic supply voltage, which closely tracks the amplitude of the transmitted RF signal. In place of a normal dc-dc converter, an ultra-high-bandwidth power modulator varies the supply voltage going to the PA to track the sample-by-sample variations in amplitude of the modulated carrier.

This significantly reduces the wasted power experienced in traditional fixed supply PAs, which are only energy efficient when they are in compression, at the peaks of the transmitted waveform. Fixed supply PAs lose up to 80% of energy (with OFDM signals) as heat since the supply voltage is far higher than required for most of the time. Tracking the RF signal closely with ET keeps the PA in compression over the whole modulation cycle, instead of just at the peaks as with the fixed supply PAs.

An important side benefit of ET is that the PA’s linearity at high power is controlled by the supply voltage, rather than the RF input signal, enabling ET to also linearize the PA. This enables full-power LTE transmission with no distortion, even for high PAPR signals, together with a boost in efficiencies to 50% and beyond. By ensuring full transmit power from the handset, ET also offers a network benefit by expanding the coverage area of a basestation by up to 60%, solving a significant network problem for operators.

ET isn’t without its own challenges. Accurately tracking the signal amplitude without causing distortion requires a power supply with bandwidth one to three times that of the channel bandwidth. Additionally, the power supply modulator has to deliver several watts of peak power and very high slew rates, without any increase in noise or letting energy conversion efficiencies dip below 80%.

Companies like Nujira have overcome these issues and are bringing single-chip ET solutions to market optimised for direct implementation in handsets. This is enabling truly global, efficient 4G performance with 2G battery life.

The Crest Of The RF Innovation Wave

ET is central to the future of the RF front end. Implementing ET can cut wasted energy in multiband PAs by more than 50%. Delivering significant improvements in power consumption and thermal management, ET is paving the way for multimode, multiband PAs that support global 4G coverage.

Poor battery life and a confusing canvas of operator-specific and region-specific handsets have greatly slowed consumer uptake of 4G phones. Operators are putting a lot of pressure on engineers to provide them with an answer to these challenges, and increasingly ET is being seen as the only viable solution for energy-efficient broadband 3G and 4G products.

Nujira expects ET to be implemented into handsets later in 2013, enabling operators and device manufacturers to overcome today’s fragmented LTE frequency band landscape and to give consumers the high-quality 4G experiences they expect.

Jeremy Hendy is the vice president of sales and marketing at Nujira. Previously, he was the marketing director of wireless USB startup Artimi, vice president of marketing for Aspex Semiconductor, and strategic technology director of Cadence’s Wireless and Multimedia business unit. He started his career with Texas Instruments and holds a first class honours degree in electronic engineering from the University of Liverpool.