There are two ongoing critical issues in wireless cellular design: power efficiency and frequency coverage.
LTE transmitters require linear power amplification because of their broadband multi-level orthogonal frequency division multiplexing (OFDM) modulation (e.g., 16QAM). These modulation techniques have a high peak to average power ratio that limits the efficiency possible in traditional linear amplifiers. Class AB linear power amplifiers (PAs) are used now.
While some improvements have been made in efficiency, it typically still lingers in the 15% to 40% range. Because most new handsets have multiple PAs to handle multiple bands, the power consumption becomes a huge part of the cell phone’s power budget, leading to more frequent battery recharges (see the figure). One solution is to provide wider bandwidth so fewer amplifiers are needed to cover needed bands or make the amplifiers more efficient.
This problem also afflicts basestations, where PAs consume the most power. Newer amplifiers use Dougherty configurations to boost efficiency and digital pre-distortion (DPD) for linearization. In envelope tracking (ET), the modulation envelope is used to vary the PA dc supply voltage so it more closely tracks the signal, resulting in class AB amplifiers that are much more efficient over a wider range of modulating signal variations. Several companies are beginning to bring ET designs to market.
Another major trend is the increasing use of gallium-nitride (GaN) transistors in the PAs rather than silicon MOSFETs or gallium-arsenide (GaAs) devices. GaN devices provide higher power at higher frequencies, and newer designs are more efficient.
There are dozens of different cellular bands, varying from country to country. Most cell phones incorporate several radios that can use any of the bands allocated to their service provider. A “world” phone that can roam anywhere would require many different bands, meaning more PAs and antennas. This problem is especially acute in LTE designs that use fragmented spectrum from the different carriers and more circuitry because of the multiple-input multiple-output (MIMO) capability. PAs and antennas generally have limited bandwidth capability inherent in their designs.
One emerging solution is the variable tuning of filters and impedance matching for antennas. Variable tuning has been difficult to implement because of circuit complexity and size. Now several companies like Peregrine Semiconductor, Ethertronics, RFMD, and Wispry are offering miniature digitally tunable capacitors that can be used to form tiny variable filters and antenna matching circuits. Both open and closed loop designs are possible.
This tunability permits one circuit or antenna to cover a wider range, greatly reducing the need for multiple components. This frees up space, reduces component count, and decreases power consumption in the handset. Tunable capacitors allow devices to operate over a wider frequency range by providing variable impedance matching and permitting antennas to be adjusted to resonance if detuned.