5G New Radio Solutions: Revolutionary Applications Here Sooner Than You Think – Part Two
This is part two of a two part article. The first part ran in the May issue of Microwave Product Digest. Click here to read it.
New Technologies Required for 5G NR —Enabling RFFE
This leads us to the discussion of what technologies, both current and new, will be required for dual connectivity 5G RFFE. At a high level, it makes sense to review the technology requirements from two separate frequency bandwidths—below 3 GHz and 3 GHz to 6 GHz—as shown in Figure 6.
When thinking about 4G LTE re-farming, it is important to understand that it will take place in relatively narrow channels using conventional PA and filter technology. While some improvements in PA output power and linearity can be expected in FDD bands below 3 GHz, sufficient performance in 4G LTE and 5G NR from the same PA duplexer paths that exist today can be generally achieved. This means that current low band technology is not only adequate even if not optimized for 5G, but with small improvements in output power and linearity it can also be optimized for 5G NR performance.
Elements of a Sub-6 GHz 5G NR RF Front-end
You may now be asking: if bands below 3 GHz can be optimized for 5G NR performance, then what are the requirements for a sub-6 GHz RFFE?
To simplify the discussion, the following section focuses on a practical example of a sub-6 GHz radio front-end with n77, n78 and n79 frequency bands. This example is intended to illustrate the important criteria for designing a 5G NR in sub-6 GHz, as shown in Figure 7. We will also discuss the impact of 5G spectrum, waveform and modulation on the constituent parts of the radio front-end module.
3GPP requirements for n77 and n79 spectrum indicate 100 MHz instantaneous bandwidth for the component carrier in the uplink. This is much more rigorous than current LTE standards that use carrier aggregation of 20 MHz channels to extend support for 40 to 60 MHz.
It is anticipated that early 5G systems will require the PA to operate in APT mode to accommodate the wider bandwidth signals. Accordingly, users can expect a 100 MHz channel when the PA is operating under APT conditions. Conversely, conventional ET is challenged to perform beyond 40 to 60 MHz. In order to extend the ET modulator bandwidth to reach 100 MHz, additional power consumption would be required, in addition to addressing amplitude/phase delay mismatch sensitivity, management of memory effects, limitations in capacitive supply loading, out-of-band Tx emissions and intermodulation into the LTE anchor band. Though there are several promising new techniques in development to extend ET to the operating bandwidth, it is projected that it will take several more years before commercialization is achieved. Designers are thus left with the challenge of delivering a better performing PA at two to three times the present state-of-the-art instantaneous bandwidth, while operating with a higher peak-to-average CP-OFDM modulation and over much larger passbands than present for 4G LTE sub-3 GHz.
Beyond the wider channel bandwidth, operators have shown significant interest in high power UE capabilities, especially as they pertain to TDD bands in the sub-6 GHz region. Currently, there is some uncertainty as to whether the bands will be 2×2 uplink (two transmitters on at the same time) or a single Tx placement, which means that the PAs will not only need to deliver industry-leading output powers compared to their 4G counterparts, but they will also have to do so over a wider bandwidth and at higher frequencies. Meeting higher output power at higher frequencies without ET modulation has given the design community problems to solve.
In order to meet the new, challenging performance requirements of wider channel bandwidth and HPUE, Skyworks engineers have developed new PA topologies that deliver linear PA performance at higher frequencies and over much greater channel bandwidth. These new architectures must be capable of significantly outperforming their LTE counterparts though under more rigorous operating conditions.
Integrated High Performance Low Noise Amplifier
When placing the sub-6 GHz modules, integrating the receive LNA functionality inside the module allows for considerable flexibility and adds value in performance. In Figure 7, there are two receive LNAs optimized for n77, n78 and n79 bands. Integrated LNAs have been proven to boost performance when overcoming system loss, especially in high frequency areas where there is generally more insertion loss due to the high frequency roll off of various RF structures.
Typically, integrated LNAs also contribute about 1.5 to 2.0 dB system noise figure enhancement, which translates directly to improved receive sensitivity when compared to alternative methods such as populating discrete LNAs at or near the transceiver.
Wideband Filter Technology
In the case of sub-6 GHz applications utilizing new TDD spectrum, legacy 4G is virtually nonexistent, except in specific regions such as Japan. While many of the defined 3GPP bands exist (B42/43/48), they have yet to be rolled out commercially in large volumes for LTE and only represent a small subset of the much larger NR band definitions. This is where we will see rapid deployment of n77, n78 and n79 RF front-end modules. We should note, however, that the passbands are significantly larger for these new 5G NR bands. For example, n77 has a passband of 900 MHz—almost 25 percent relative bandwidth, which is twice as large as the 5 GHz Wi-Fi band —and n79 has a passband of 600 MHz. In both instances, we find that conventional acoustic filters are not well-suited for these extremely wide passbands.
There are additional complexities that will determine the extent of 5G NR wideband filter requirements. For example, one can derive a simple filter response if we assume an ideal environment with a separate high band antenna and no coexistence requirements. On the other hand, if we consider a more complex radio environment, such as a multi-radio atmosphere with simultaneous Wi-Fi transmission, you will see the filter requirements become much more strenuous.
Thus, it is important to take note of the radio environment, antenna topology and coexistence requirements in order to specify the optimum filter. In other words, the filter design and antenna topology into which the FEM will be subsequently mated must be customized to the specific use case or application. Skyworks has the expertise to tailor 5G NR filters to accommodate either end of these extreme use cases.
Antenna Outputs and Fast Sounding Reference Signal (SRS) Hopping
Antenna configurations will play a vital role in the mainstream 5G products. While market requirements will slowly become clear over the next 18 months, there is already some uncertainty as to which of the optional features will be supported.
One feature is the fast hopping sounding reference signal (SRS), which uses the transmitter to send a series of known symbols across all of the downlink receive antennas in the UE in order to better calibrate the MIMO channel and improve the downlink SNR. This process is key to enhanced MIMO and beam forming operation. SRS carrier switching (SRS-CS) was recently introduced in LTE Release 14 to assist the eNodeB (eNB) in obtaining the channel state information (CSI) of secondary TDD cells in TDD LTE CA scenarios. Prior to Release 14, only the primary cell benefits from SRS UE transmissions, and therefore downlink transmissions on secondary cells are done without prior knowledge of the CSI (Figure 8).
SRS transmit switching (SRS-TS) allows UE to route its SRS transmissions to all other available antenna ports. Assuming channel reciprocity holds, as should be the case for TDD operation, this feature enables gNB, or 5G NR base station, to estimate CSI on secondary downlink “only” cells. Applying that concept to multi-user MIMO (MU-MIMO) offers yet another network performance enhancement, which in turn will enhance the 5G consumer experience.
Figure 9 provides an example of fast SRS hopping transmissions to any receive downlink antenna port for a UE architecture supporting 1 Tx/4 Rx operation. This scheme requires an RF switch to route the UE transmitter chain to each of the remaining 3 Rx antenna ports.
In 5G NR, the same architecture is relevant—SRS-TS allows gNB to evaluate cell CSI, assuming channel reciprocity applies. This is essential for MU-MIMO performance and MIMO performance at high frequency, particularly as the channel coherence time is short and only fast SRS hopping can provide sufficient MIMO channel estimation.
Implementation Scenarios: What Will a 5G Enabled UE Look Like?
Now let’s take a look at what we envision a typical implementation of new 5G NR features would be in several different usage cases.
Implementation in FDD LTE Re-farming Bands Below 3 GHz
The previous section covered greenfield operation in new TDD bands called the sub-6 GHz range. In this section, we take a quick look at a different usage case—specifically, how some operators plan to re-farm their LTE bands into 5G NR. There are two key differences when running 5G NR modulation through an existing 4G PA path: (1) the filter bandwidth and isolations may have to change on the receive side and (2) on the transmit side, the PA may require some incrementally increased linearity and power capability.
Initially, LTE re-farming for 5G NR can be accomplished using the low, mid and high band power amplifier modules with integrated duplexer (PAMiD) structures currently utilized in the market. However, as we get closer to commercial network rollout in 2020 and beyond, improvements will need to be made in the PAMiDs to accommodate for both 4G and 5G LTE re-farmed operations.
Putting It All Together: What Does a Dual Connectivity 4G LTE/5G NR RF Front-end Look Like?
In Figures 10 and 11, Skyworks illustrates a possible solution for dual connectivity smartphones. There are many ways to achieve the same goal. However, using a 4G core front-end with the addition of 5G NR modules to support overlaid 5G performance and dual connectivity offers a simple, straightforward solution. Figures 10 and 11 show conventional PAMiD devices in the primary transmit path and diversity receive components on the diversity antenna side. This is standard implementation of the core 4G content of a dual connectivity RFFE.
In order to achieve full performance sub-6 GHz UE, there are additional placements required for the transmit capability in bands n41, n77, n78 and n79 to support 2×2 uplink MIMO. In the new n77, n78, and n79 bands in particular, this means an additional 5G NR PAMiD module, as well as two additional diversity receive components needed to support downlink 4×4 MIMO capability.
A Look Forward to 5G Commercial Networks
How might 5G requirements evolve as we get closer to commercial implementation?
As mentioned earlier, the standards, operator requirements, device manufacturer plans and chipset architectures are being finalized throughout 2018 and 2019. In the interim, the industry is moving forward with demonstrating practical 5G compliance and testing methodologies and projects in order to achieve commercialization in the second half of 2019 and early 2020 timeframe.
This is a very compressed timeline to establish all the components of the ecosystem, as well as to standardize equipment, install small cells, test devices and chipsets and align smartphone OEMs in order to deliver groundbreaking solutions.
While we anticipate changes, we are confident that the course defined in this article is one that can deliver pragmatic, yet groundbreaking performance solutions to the market in a timely manner.
Summary: Enabling a New Era of Disruptive Communications Technology
Throughout many conversations with mobile network operators, UE device manufacturers and chipset partners, it is clear that 5G will be here sooner than many predict. As with previous generations, the transition to a new technology presents opportunities as well as challenges. With decades of experience spanning multiple generations of wireless standards, Skyworks possesses the technology breadth and expertise to meet these challenges and deliver increasingly sophisticated solutions to advance both the vision and promise of 5G.
From the early days of 2G communication to the more digital wireless standards of 3G WCDMA and 4G LTE, Skyworks has been at the center of innovation and manufacturing scale with each generation (see Figure 12). 5G will be no exception. 5G will raise the bar on system performance and push improvements in size, integration, coexistence and modulation distortion in order to meet design criteria for expanding markets such as enhanced mobile broadband (eMBB), cellular vehicle-to-everything (C-V2X) and low latency communications. Skyworks, via its Sky5™ unifying platform, is working diligently to ensure the ecosystem and platforms are in place to ensure a seamless transition to this new era of exciting and previously unimagined applications.
“Skyworks” along with “Star Design” logo are registered trademarks of Skyworks Solutions, Inc.
All third party trademarks (including names, logos and icons) referenced by Skyworks Solutions, Inc. (“Skyworks”) remain the property of their respective owners. Unless specifically identified as such, Skyworks’ use of third party trademarks does not indicate any relationship, sponsorship, or endorsement between Skyworks and the owners of these trademarks. Any references by Skyworks to third party trademarks are to identify the corresponding third party goods and/or services and shall be considered nominative fair use under the trademark law.
 R4-1706709. 2Tx UE for NR. 3GPP TSG-RAN WG4 Meeting NR #2, June 27–29, 2017. Qingdao, China.
 38.101-1-f20 (Rel 15.2). NR; User Equipment (UE) radio transmission and reception; Part 1: Range 1 Standalone. Retrieved June 14, 2018, from www.3gpp.org/DynaReport/38101-1.htm.
 38.101-2-f20 (Rel 15.2). NR; User Equipment (UE) radio transmission and reception; Part 2: Range 2 Standalone. Retrieved June 14, 2018, from www.3gpp.org/DynaReport/38101-2.htm.
 38.101-3-f20 (Rel 15.2). NR; User Equipment (UE) radio transmission and reception; Part 3: Range 1 and Range 2 Interworking operation with other radios. Retrieved June 14, 2018, from http://www.3gpp.org/DynaReport/38101-3.htm
For more information about our solutions, please visit us at www.skyworksinc.com/5g