Why 5G is going to over-the-air testing

Cables and connectors may become artifacts of the past for communication electronics as the world moves to 5G standards.

SHERI DETOMASI, KEYSIGHT TECHNOLOGIES, INC.
5G is rapidly approaching and there is a lot of talk about over-the-air (OTA) testing – that is, testing without the DUT physically connected to the test equipment. I’ve seen test vendors show off their equipment with multi-element antennas attached, or even fully operational anechoic test chambers featuring automated OTA testing of wireless devices. Equipment vendors and operators now promote OTA test solutions as well. This scenario differs dramatically from 4G where most tests used cables. So, what is driving the emphasis on OTA testing in 5G?

First a little background. The 5G vision set forth by IMT-2020 is amazing. Conceived by the International Telecommunication Union Radiocommunication Sector, this term stands for International Mobile Telecommunication system with a target date set for 2020 (for standardization). It opens many possibilities for consumers, the environment, humanity, health and safety. Virtually every industry will be transformed, and new ones will emerge. Three defined use cases represent the foundation for the 5G specifications: enhanced mobile broadband (eMBB) to support extreme data rates, ultra-reliable low-latency communications (URLL) near-instant communications, and massive machine-type communications (mMTC) for massive IoT interconnects.

It will take new technologies to enable the 5G vision. The 5G New Radio (NR) specification adds new operating bands in both sub-6-GHz and millimeter-wave (mmWave) frequencies to extend the available spectrum. 5G NR identifies almost 10 GHz of new spectrum in sub-6 GHz and up, to mmWave frequencies that will be implemented on a country-by-country basis.

FR1 (Frequency Range 1): 400 MHz to 6 GHz adds 1.5 GHz of new spectrum in frequency bands: 3.3 – 4.2 GHz, 3.3 – 3.8 GHz, 4.4 – 5 GHz

FR2 (Frequency Range 2): 24.25 to 52.6 GHz adds 8.25 GHz of new spectrum in frequency bands: 26.5 — 29.5 GHz, 24.25 — 27.5 GHz, 37 – 40 GHz. Initial target frequency bands are 28 GHz and 39 GHz.

FR2 mmWave frequencies up to 52.6 GHz can have channel bandwidths up to 800 MHz when carrier aggregation is used to combine multiple component carriers. This additional spectrum is essential to enabling 5G’s promise of extreme data rates of 20 Gbps in the downlink (DL) and 10 Gbps in the uplink (UL).

While mmWave frequencies enable more bandwidth, they also expose signal propagation issues that are not a problem at sub-6-GHz frequencies. Signal propagation issues such as increased path loss, delay spread, and blockage make it more difficult to establish and maintain a wireless communications link between a mobile device and base station.

At mmWave frequencies, the placement of a user’s hand on a mobile device or the orientation of their body can significantly degrade the radio link performance. Signals also experience more attenuation going through different types of materials, which can limit outdoor-to-indoor coverage. Non-line-of-sight (NLOS) scenarios can also incur delay spread.

A phased array antenna is formed by an array of smaller antenna elements.

To overcome these signal propagation issues, 5G radio systems will utilize multiple antennas on base stations and mobile devices to implement spatial diversity and beam steering techniques to reliably direct narrow beams in a specific direction.

Beam properties at different distances from the antenna array.

In sub-6-GHz frequency bands, massive MIMO (multiple-input, multiple output) technologies will be deployed to boost cell capacity and realize the throughput envisioned by eMBB use cases. Massive MIMO will use a much greater number of antenna elements on the base station directed to multiple user devices simultaneously in the same frequency and time resource, also known as multi-user MIMO (MU-MIMO). With the large number of antenna elements, massive MIMO antenna designers will be challenged with the ability to visualize, characterize, and validate the antenna array beam patterns.

WHY OTA TESTS?

In that 5G will utilize multiple antennas for massive MIMO and beam steering, consider what’s driving the need for OTA tests:

Increasing density of multi-element antenna arrays – Phased-array antennas are becoming a preferred technique for mobile communications. These antennas can use changes in relative phase and amplitudes of a signal applied to each antenna element to create narrow beams and dynamically steer the beams in a desired direction. This is also known as beam steering.

Phased-array antennas will be used in sub-6-GHz frequency bands to boost capacity and in mmWave frequency bands to overcome path loss issues. 5G NR (New Radio) Release 15 specifies up to eight layers, or streams of data, in the downlink, and up to four layers in the uplink. While this is not an improvement over 4G LTE-Advanced, expectations are that 5G base stations may employ hundreds of antennas for massive MIMO, and devices will implement many more antennas than are in production today.

Due to the number of antenna elements configured as narrow beams, it will be difficult to fully characterize and validate beam performance. Designers will need to measure beam patterns both in 2D and 3D and understand beam width, side lobe levels, null depths, and symmetry. These types of tests must use OTA test methods.

Ever-smaller mmWave components — Higher frequencies imply shorter wavelengths. Multi-antenna arrays at mmWave frequencies will help overcome signal propagation issues and deliver directional antennas with higher gain. With shorter wavelengths, antenna elements can be spaced more tightly, resulting in extremely compact arrays. Many vendors even opt to develop arrays that integrate into ICs. Of course, highly integrated ICs have no place to probe and no place to put connectors. A consequence of this integration is it has become impractical to use traditional RF connectors between the radio circuit and the antenna, bringing the need for OTA tests.

HOW ARE OTA TESTS PERFORMED?

OTA tests typically take place in either the near-field or far-field regions of the antenna array. The characteristics of the transmitted electromagnetic (EM) wave change depending on the distance from the transmitter. As the signal propagates from the antenna array, the signal becomes more developed. The amplitude of the peaks, sidelobes, and nulls in the radiation pattern evolve towards the far-field pattern.

Close to the antenna, typically on the order of a few wavelengths or less, resides the reactive near-field. The far-field, or Fraunhofer distance, begins at 2D²/λ, where D is the maximum diameter of the radiating elements and λ is the wavelength. Some communication systems, such as near-field communications (NFC) or radio-frequency identification (RFID), use the near-field region for communication. However, 5G cellular communication links must be evaluated using far-field assumptions.

Estimated far-field distance and path loss for different radiating apertures.

Measurements such as radiated power can take place in the near-field. However, the far-field beam pattern is not fully formed in the near-field. Some near-field scanning techniques, using a Fourier transform of the near-field pattern, can be used to predict the far-field pattern. However, measurements in the reactive near-field are not as accurate because the receiving antenna can interact with the transmitting antenna and degrade measurement results.

We can calculate the far-field distance for a typical 5G mmWave device. Assume a 15-cm radiating antenna element (D) operating at 28 GHz; from the above equation, it would have a far-field distance of 4.2 m and a path loss of approximately 73 dB.

This kind of distance introduces new challenges in design and test. RF parameters such as transmitted power, transmit signal quality, and spurious emissions are measured in radiated transmitter tests. Such tests are more difficult as the far-field distance lengthens and the path loss grows. To make matters worse, path loss deteriorates as the radiating element dimensions (D) get larger or the frequency rises.

It takes different instrumentation setups to gauge RF parameters as designs progress from early R&D through conformance and manufacturing tests. During the prototyping phase, designers must characterize the performance of the chipsets, antennas, and devices in a controlled over-the-air environment. Engineers will need to characterize and evaluate their designs to meet the minimum requirements specified by the 3GPP (3rd Generation Partnership Project, the group developing and maintaining 5G standards) before releasing their device to the market.

In 4G, radiated tests are required for safety (Specific Absorption Rate – SAR), electromagnetic compatibility (EMC), and more recently to validate MIMO throughput. Most other tests take place in a cabled or conducted environment.

Moving to 5G, the low-frequency tests will be like 4G tests. But for mmWave, the following tests now must take place using OTA methods:

RF performance – Minimum level of signal quality
Demodulation – Data throughput performance
RRM – Radio resource management – Initial access, handover and mobility
Signaling – Upper layer signaling procedures
Manufacturing test – Calibration and validation of performance

Standards committees have yet to define many of these tests. RF radio transmission and reception requirements for NR user equipment (UE) and base stations, and conformance tests, for example, will be specified in the 3GPP 38 Series.

COMPARING OTA TEST METHODS

OTA tests will be critical to developing, validating, and commercializing 5G NR devices. A typical OTA test will involve an anechoic chamber, different probing techniques, and test equipment to generate and analyze the radiated signals in a spatial setting. The anechoic chamber provides a non-reflective environment with shielding from outside interferences so radiated signals of known power and direction can be generated and measured in a controlled environment.

As defined by the 3GPP, the test setup for the traditional DFF test method (top), the IFF test method based on a CATR (middle), and the near-field with transform (NFTF) test method (bottom) for EIRP/TRP measurements (used with permission).

To date, the 3GPP has defined three permitted test methods: direct far-field method (DFF), indirect far-field method (IFF), and near-field to far-field transform (NFTF). In the DFF method, the DUT mounts on a positioner that rotates in azimuth and elevation to enable measurement of the DUT at any angle on the full 3D sphere. The range length of the chamber is determined by the Fraunhofer far-field distance mentioned earlier. The IFF test method is based on a compact antenna test range (CATR) and uses a parabolic reflector to collimate the signals transmitted by the probe antenna to create a far-field test environment in a much shorter distance than the DFF method. The NFTF method samples the phase and amplitude of the electrical field in the near-region and uses a Fourier transform to predict the far-field pattern.

The best test method depends on the radiating DUT antenna size and configuration. While the IFF CATR method can be used for the three current DUT categories identified by 3GPP, the excessive path loss with larger radiating elements limit the DFF method to DUTs having radiating antenna elements smaller than 5 cm.

Both DFF and IFF methods can be used for RF parametric tests to characterize beam patterns and validate beam steering. The main differences are in the required size of the chamber and the associated path loss. The CATR method is most flexible with DUT size and frequency requirements, and the small chamber involved can be used in lab environments.

An example setup to reconstruct realistic channel conditions for the signals transmitted/received by a base station or device emulator.

To understand device performance in real-world conditions, developers test a device’s end-to-end system performance while including impaired signals. This can take place using a PROPSIM 5G channel emulator. A channel emulator simulates real-world signal impairments including path loss, multi-path fading, delay spread, and Doppler shift.

To ensure accurate and repeatable measurements through a design cycle, new OTA test methods are being studied and approved by 3GPP. Direct far-field, indirect far-field, and near-field to far-field transformation has been approved by 3GPP to date. As 5G goes mainstream, OTA test methods will be critical for moving from R&D to design verification, through conformance and into manufacturing.

References
Keysight Technologies Inc., 5G solutions

More information about OTA challenges and test methods can be found in the OTA Test for millimeter-wave 5G NR Devices and Systems White Paper

3GPP TR 38.310 technical report defining DFF, IFF, and NFTF test methods