Growing demand for mobile video and other applications is driving an accelerated timeline for delivering 5G to smartphones and other mobile devices. In March 2017, the 3GPP standards group approved a new standard, for the 5G non-standalone (NSA) radio. This standard enables mobile operators to deliver 5G enhanced mobile broadband (eMBB) service by leveraging their existing 4G infrastructure. As a result, 5G mobile deployments are expected to start as early as 2019, and initial 5G implementations are expected to use frequencies between 3.3 and 3.8 GHz.
Figure 1 5G vs. 4G performance capabilities.
In mobile broadband, 5G will complement 4G rather than replacing it. LTE Advanced and LTE Advanced Pro networks using carrier aggregation are expected to provide peak data rates up to roughly 1 Gbps, which is adequate for many uses. 5G will add much higher peak network speeds—up to around 10 Gbps—to support bandwidth-intensive applications such as live video and augmented reality (see Figure 1). Unlike the 5G standalone (SA) specification, 5G NSA uses the existing 4G LTE radio and core network as an “anchor” for control and management, together with a 5G data carrier.
Supporting 5G creates significant challenges for the mobile device RF front-end (RFFE), including new frequency bands, massive bandwidth, high output power and both 4G and 5G waveforms. The Qorvo QM19000, the world’s first 5G mobile RFFE, is designed to meet these challenges. As shown in Figure 2, the QM19000 integrates the power amplifier (PA), low noise amplifier (LNA), transmit/receive (Tx/Rx) switch and wideband filter into a single module that supports up to 400 MHz bandwidth over the 3.3 to 4.2 GHz frequency range.
Figure 2 Functionality of the QM19000 RFFE module.
The RFFE power efficiency and linearity requirements for 5G are particularly complex, due to the need for backward compatibility in regional applications where the 3.5 GHz bands have already been utilized for LTE. For example, 5G signals are much wider than those used in 4G; a typical goal for 5G Tx signal bandwidth is 100 MHz, compared with the 20 MHz maximum provided by a single 4G LTE carrier. For 4G, envelope tracking (ET) is widely used in mobile front-ends to maximize power efficiency; it cannot be used for 5G, because today’s envelope trackers support a maximum bandwidth of 40 MHz. Instead of ET, average power tracking (APT) mode (i.e., fixed voltage) must be used for 5G. The PA must provide both high linear efficiency for 5G transmissions as well as high saturated efficiency when using ET for 4G signals. The QM19000 supports sophisticated power management that enables the module to switch between ET and APT mode.
Adding to the challenges, 5G specifications define two alternative waveforms: cyclic-prefix orthogonal frequency division multiplexing (CP-OFDM) and discrete Fourier transform spread OFDM (DFT-S-OFDM). CP-OFDM offers very high spectral packing efficiency in resource blocks (up to 98 percent) and good support for multiple-input-multiple-output (MIMO). It is likely to be used when an operator’s priority is to maximize network capacity, such as in dense urban environments. The combination of CP-OFDM and massive bandwidth generates much higher peak-to-average power ratios (PAR) than with LTE, requiring greater back-off in the PA to avoid exceeding regulatory limits. As a result, the PA must offer high-power efficiency and linearity at a wide range of power levels. The RFFE also must support DFT-S-OFDM, the same waveform used for the LTE uplink, which provides less efficient spectral packing but greater range.
The 5G RFFE also requires support for power class 2, a specification for higher output power to overcome greater propagation losses at high frequencies. Originally approved for specific 4G bands, power class 2 doubles the output power at the antenna to 26 dBm. With 5G SA, power class 2 is a baseline requirement across all new bands.
To meet all the Tx requirements for both 4G and 5G, including saturated and linear efficiency, the QM19000 uses a two-stage PA manufactured using Qorvo’s GaAs HBT5 process. In addition to improved efficiency, HBT5 offers higher gain than other technologies, such as SiGe, at the high frequencies used for 5G. HBT5 PAs also have the thermal performance required to support higher power output and use Qorvo’s CuFlip copper-bump packaging technology to efficiently dissipate heat. Even with the HBT PA, additional amplification is needed to meet specific requirements for higher output power, such as power class 2. Additional gain may also be required when using CP-OFMD, when transceiver drive levels are expected to decrease by up to 3 dB. The QM19000 includes an additional variable gain amplifier to meet these needs.
While supporting 5G performance requirements, the QM19000 also helps smartphone manufacturers accommodate complex RF functionality within tight space constraints. For example, MIMO is key to delivering the high data rates promised by 5G. Device manufacturers are expected to use 4 × 4 MIMO for downlink, and some implementations may additionally provide 2x MIMO for the uplink. This means manufacturers need to fit even more RF chains into the essentially fixed space allocated to the RF content within the smartphone. Highly integrated modules such as the QM19000 will be essential to achieving this goal.
Over time, 5G will address a range of different use cases, each presenting new RF challenges. The QM19000 is one element in Qorvo’s expanding portfolio of 5G solutions designed to meet those challenges.
Qorvo Inc.
Greensboro, N.C.
www.qorvo.com