5G New Radio Signal Design

By Gent Paparisto, Joel Kirshman, and David Vye

The CP-OFDM modulation scheme will have many options. Variable subcarrier spacing (SCS) will be 3.75, 15, 30, 60, 120, 240, 480 kHz with sub-1 GHz of 15 and 30 kHz and sub-6 GHz of 15, 30, and 60 kHz, and above 6 GHz of 60 and 120 kHz. There will be normal and extended CP options. Slot duration will be 0.5 ms, subframe curation will be 1 ms, and frame duration will be 10 ms. Component carrier (CC) bandwidth will be minimum 5 MHz and maximum 100 MHz for sub-6 GHz and for greater than 6 GHz a minimum 50 MHz and maximum 400 MHz. The maximum number of programmable subcarriers could be up to 100 percent. There are some discussions of the 3GPP radio specification group (RAN4) pushing up to 92-94 percent carrier occupancy with FFT4096 as mandatory and FFT8102 being considered. Carrier aggregation will be a maximum of 16 carriers for RAN1 and a maximum of 32 carriers for RAN2. Modulation will be QPSK, 16 QAM, 65 QAM, and 256 QAM with UL support π/2-BPSK.

Table 1 shows the 5G NR range of values allowed that have been decided by RAN4 for the purpose of radio specifications to be made in the near future for EVM, adjacent channel power ratio (ACPR), and more.

Table 1. 5G NR range of values allowed by RAN4.

A simulation tool used by PA or DUT designers must have the ability to create these signals efficiently with simple configuration options. VSS provides these component designers with preconfigured test benches that can be modified to their needs by “dialing in” the appropriate parameters, relying on the software to accurately generate the signal, and make the desired measurements.

VSS Verizon 5G Carrier Aggregation Test Bench
VSS provides test benches for generating Verizon 5G (VZ5G) signals currently under development from the 5G Technical Forum and lead by Verizon. Furthermore, as shown in Figure 7, VSS offers test benches that implement the VZ5G carrier aggregation capabilities, which can simultaneously generate up to eight signals at channels spaced 99 MHz or wider, and perform EVM, ACPR, or other measurements.

Figure 1: VSS LTE carrier aggregation for Verizon 5G work bench.

Drilling down into one of the signal generators, it can be seen that the signal generator is built from basic building blocks, allowing system designers to modify as needed and keep current with any changes in specifications. Figure 8 shows details for the VZ5G base station transmitter, including scrambling, modulation mapping, layer mapping, and precoding. As opposed to using “black box” designs, developers can go inside the VSS VZ5G signal generation/receiver blocks and change their parameters, or even their architecture, to perform various trade-off analysis or keep up to date with the specification changes. Because the NI AWR software team closely follows the work of the 5G Technical Forum, VSS test benches are updated accordingly.

Figure 2: VZ5G base-station transmitter details, including PDDCH, PDCCH, and synchronize/reference signal per VZ5G.

Although most specifications for VZ5G are the same as for LTE, scrambling is different. The transmissions contain the physical-downlink shared channel for data (PDSCH), physical-downlink control channel (PDCCH), and sync/reference signals, all of which are different than LTE. These signals are combined into a frame with a specific architecture, which is different in VZ5G. Hence, the LTE frame assembler could not be used a new one has been created. The OFDM modulator is similar to LTE with the main difference being the subcarrier spacing.

Many aspects of the VZ5G specifications are similar to LTE, but there are also quite a few differences. The transmission contains a physical-downlink shared channel for data (PDSCH), a physical downlink control channel (PDCCH), and sync/reference signals, all of which are different than LTE. These signals are combined into a frame with a specific architecture, which is different in VZ5G. Hence, the LTE frame assembler could not be reused and a new one has been created. The OFDM modulator is similar to LTE with the main difference being the subcarrier spacing.

When simulated in VSS, the environment enables the user to look at the resulting signals that are generated (Figure 9). The ACPR measurement interface is similar to instrumentation, where the user is able to define the carrier frequency, channel bandwidth, adjacent channel offset and corresponding bandwidth, as well as simulate swept power and simultaneously monitor ACPR measurements.

Figure 3: Verizon 5G base station example analysis of ACPR.

This particular project supports EVM measurements where designers can look at the IQ plot as a function of increasing output power, where power saturation will introduce signal distortion above the 17.5 dBm level for the particular amplifier used in this project. Figure 10 shows a 64-QAM (data) and a QPSK signal (control) in the background (green trace). An EVM trace is indicated for a simulation that was run without a device under test (DUT). The DUT was then added and the direct impact of EVM on the DUT can be seen. The DUT could be just an amplifier or a whole transmit or receive chain, inclusive or exclusive of phase noise-there is a great deal of flexibility in the environment for what the user might want to simulate.

Figure 4: VZ5G base station example analysis of EVM.

The current 5G non-standalone NR standard defines different subcarrier spacings and carrier frequencies. One of the additions to the new specification is the use of 4096-point FFT. This means there is a large variation in what types of signals will be generated in NR.

VSS supports signal generation at any frequency, from below 6 GHz all the way up to mmWave and all the formats that are currently specified by 3GPP for 5G NR. With the VSS test bench shown in Figure 11, the user can simulate a particular NR signal configuration, define the modulation type, place it at the frequency of choice, pass it through a nonlinear device, and monitor spectrum and any other measurements.

Figure 5: VSS test bench environment for 5G (256 QAM, carrier frequency of 3.5 GHz).

5G NR Frequency-Band Options
5G deployment will be prioritized in three band types per the RAN4 radio performance protocol:

• LTE re-farming bands: 1, 3, 7, 8, 20, 28, 41, 66, 70, 71
  SCS: 15 kHz, 30 kHz and 60 GHz (for bands above 1 GHz)
• New sub-6GHz bands: 3.3-4.2 GHz and 4.4-4.9 GHz
  SCS: 15 kHz, 30 kHz and 60 kHz (for bands above 1 GHz)
• mmWave bands: 24.25-29.5 GHz, 31.8-33.4 GHz, and 37-40 GHz
  SCS: 60 kHz and 120 kHz

Band-specific aspects of the system can be easily addressed by the VSS user to support channel bandwidths that will vary (BS/UE could also have different bandwidths). Primary synchronization signal (PSS)/ secondary synchronization signal (SSS)/physical broadcast channel (PBCH) synchronization, and broadcast block locations will vary and will need a flexible frame assembler. Supported MIMO modes will also likely vary for different bands-discussion is ongoing at 3GPP.

5G Simulation Roadmap
The NI AWR software team continues to develop 5G capabilities for the evolving 5G ecosystem including 5G NR, VZ5G TF, 5G-channel model, and MIMO operations. 5G NR development in VSS consists of CP-OFDM modulation and various SCS settings. There will be a flexible frame assembler/disassembler, a phase-tracking reference signal (PT-RS) and RX phase-noise compensation for both mmWave and wideband operation, and PSS/SSS/PBCH capability, depending on operating frequency and SCS.

For the VZ5G Technical Forum, the current implementation in VSS will be updated according to standardization progress. The current VSS signal channel model (SCM) in the 5G-channel model will be updated to the final configuration defined in 3GPP. Finally, VSS will continue to update MIMO signal generation and MIMO decoding to support any updates in the specifications.

NB-IoT in VSS
New technology for IoT was introduced in 3GPP Release 13. It is designed to coexist with global system for mobile communications (GSM), general packet radio service (GPRS), and LTE. Three modes of operation are defined: standalone, in-band, and in-guard-band. Peak data rates for download will be 226.7 kbps and 250 kbps for upload. Bandwidth will be 200 kHz for standalone and 180 kHz for in-band and in-guard-band. Modulation will be π/2-BPSK or π/4-QPSK with phase continuity between symbols. Downlink will be OFDMA with 15 kHz subcarrier spacing and uplink will be SC-FDMA with 15 kHz subcarrier spacing and single tone with 3.75 kHz and 15 kHz. NB IoT will also have half duplex frequency-division duplexing (FDD).

VSS supports all three operating modes with both signal generation and receiver functionality. Test benches in the current version of the software provide examples to analyze NB-IoT performance, NB-IoT co-existence with LTE, and measure receiver sensitivity and throughput (Figure 12).

Figure 6: Four example VSS NB-IoT projects: LTE and NB-IoT uplink coexistence test bench, NB-IoT uplink test bench, NB-IoT in-band uplink test bench, and NB-IoT uplink test bench.

Conclusion
5G NR is the wireless standard set to become the foundation for the next generation of mobile networks and new candidate technologies for modulation, antennas, spectrum, and higher density are being developed to support the new standard. VSS system simulation software offers a 5G modulation waveforms library, inclusive of test benches and phased-array features to study key performance metrics such as EVM, ACPR, BER and CCDF, giving users access to current 5G standard and candidate signals, measurement (simulation) test benches, and re-configurable 5G Tx/Rx sub-systems for further modification and/or architecture development.

Gent Paparisto is a product manager for RF Systems at NI’s AWR Group; Joel Kirshman is a marketing researcher for NI’s AWR Group.