5G Massive MIMO System Architecture and Test Technology

1 Introduction

Massive MIMO (massive antenna) technology is one of the key technologies of 4.5G/5G, and global communications operators are paying close attention to Massive MIMO technology. China Mobile and Japan's SoftBank have already launched TD-LTE Massive MIMO technology. Operators such as China Unicom, China Telecom, and Telkomsel have completed FDD Massive MIMO field tests. In my country's first phase of 5G trials, Massive MIMO was used as a key technology, and five manufacturers including Huawei, ZTE, and Ericsson participated in the trials. 3GPP has included support for Massive MIMO as one of its important features since version R13.

Massive MIMO technology uses a large number (such as 64/128/256, etc.) of array antennas on base station transceivers to achieve greater wireless data traffic and connection reliability. Compared with previous single/dual-polarization antennas and 4/8-channel antennas, massive antenna technology can improve the utilization efficiency of spectrum and energy through different dimensions (spatial, time, frequency, polarization, etc.); 3D shaping and channel estimation technology can adaptively adjust the phase and power of each antenna array, significantly improve the beam pointing accuracy of the system, and concentrate the signal strength on specific pointing areas and specific user groups. While enhancing user signals, it can significantly reduce self-interference and neighboring interference in the cell, which is an excellent technology to improve the carrier-to-noise ratio of user signals.

How to evaluate Massive MIMO technology, what kind of test indicators and test methods to use, and how to measure Massive MIMO technology fairly and efficiently? These are also issues that the current communications technology industry is very concerned about.

2. Massive MIMO system architecture

The active antenna base station architecture supporting Massive MIMO is represented by three main functional modules: RF transceiver unit array, RF distribution network and multi-antenna array.

The RF transceiver array contains multiple transmitting units and receiving units. The transmitting unit obtains baseband input and provides RF transmission output, which will be distributed to the antenna array through the RF distribution network. The receiving unit performs the opposite work of the transmitting unit. The RDN distributes the output signal to the corresponding antenna path and antenna unit, and distributes the input signal of the antenna to the opposite direction.

The RDN may comprise a simple one-to-one mapping between transmit units (or receive units) and passive antenna arrays. In this case, the radio frequency distribution network would be a logical entity but not necessarily a physical entity.

Antenna arrays may include various implementations and configurations, such as polarization, spatial separation, etc.

The physical locations of the RF transceiver array, RF distribution network, and antenna array may differ from the logical representation in the following figure, depending on the implementation.

Figure 1. Active antenna base station architecture supporting Massive MIMO

3. Massive MIMO Testing Technology

3.1 Challenges of the Evolution of Antenna Systems to Test Technology

With the modernization of antenna systems, especially the evolution of 5G, the integrated base station active antenna system (AAS) has gradually become the mainstream. The number of channels is increasing, the connection method of active antennas will be simplified, the RU and antenna will be highly integrated, and the RF indicators will no longer be limited to the traditional RU conduction test. OTA testing will become the direction of future test evolution, but it will also bring great testing challenges.

Table 1. Challenges of antenna system evolution to test technology

3.2 Test signal modulation

Figure 2. Test signal modulation

Active antennas work under various service carrier states to achieve network coverage. To truly test the performance of active antennas, the test system needs to have the following test capabilities:

1. The test system needs to support the amplitude and phase test of the service signal, especially the test of the large bandwidth signal;

2. The directional pattern test signal mode needs to be discussed and defined.

3.3 Antenna Beam Diversification

Figure 3. Massive MIMO antenna network coverage diagram

In scenarios where antenna beam radiation characteristics tend to be complex:

1: How to accurately evaluate the pointing accuracy, side lobes, beam width, etc. of antenna business beams;
2: How to select multi-beam test scenarios;
3: The test efficiency of multi-beam antennas;
4: How to evaluate the coverage performance of multi-beam antennas through two-dimensional radiation characteristics.

Testing suggestions:

1: It is necessary to evaluate the index requirements of active antennas, especially Massive MIMO antennas, on two main surfaces; it is necessary to study and define the 3D radiation index requirements;
2: Evaluate the multi-beam radiation performance under real business signals and establish a test case set.

3.4. High-frequency communication antenna band

High-frequency (millimeter wave) coverage has always been a difficult problem in the industry, but Massive MIMO can solve this problem well. As an extended frequency band of 5G, it provides capacity guarantee.

With the same number of antenna units, the higher the frequency, the shorter the coverage distance. High-frequency millimeter waves have a natural disadvantage in coverage, but in theory this can be compensated by increasing the number of antennas. As the frequency band increases, in order to achieve the same coverage distance, the number of antenna units needs to be increased, which means an increase in antenna costs. Therefore, reducing antenna costs has become one of the key issues for 5G multi-antenna technology.

As one of the key technologies for 5G evolution, high-frequency Massive MIMO antenna has several key features: high frequency, large bandwidth, and ultra-large-scale array antenna.

These key features place new demands on testing:

a) The radiation indicators of high-frequency antennas need to be re-analyzed and redefined;
b) The test site and instruments need to support the testing of large-aperture ultra-high frequency antennas, especially the testing of OTA characteristics;
c) The test instruments need to support the testing of ultra-high frequency and ultra-wideband signals.

3.5. Radio Frequency Index Test Air Interface

With the development of antenna integration, especially Massive MIMO antennas, RF conduction radio frequency indicators have radiation directionality and a large number of channels. How to test RF indicators is a huge challenge currently encountered. There is no clear technical approach at present, and the 3GPP standard is also under technical discussion. One of the current directions is to conduct air interface testing, but how to define the air interface performance of these RF indicators and how to test them are currently difficult problems in the industry.

At present, the 3gpp R13 standard clearly defines EIRP and EIS for RF index air interface testing, and other air interface indicators have been analyzed in the recent RAN R14 standard. At present, this part is very complicated, and all parties are studying it. There is no clear conclusion on how to conduct air interface testing for these RF indicators.

Currently it is mainly divided into two parts:

a) In-band indicators: At present, if the antenna performance is known, it can be evaluated through the existing OTA test solution.
b) Out-of-band indicators: The out-of-band performance of the antenna is unknown, and the very wide frequency points out of the band are a huge challenge for air interface testing!

3.6 3D Beamforming Characteristics

3D-beamforming based on accurate channel estimation may have limitations in describing beam characteristics with traditional multiple 2D cross sections. As shown in the following figure, the traditional E-plane and H-plane cutting cannot reflect the beam sidelobe distribution characteristics. In addition, the service beams of Massive MIMO antennas change with users, so it is almost impossible to traverse and test all beam scenarios. In actual testing, it is recommended to select typical service scenarios for testing.

Compared with traditional antenna coverage, the service beam of Massive MIMO antenna may be narrower, and its pointing accuracy directly affects the network coverage performance. Therefore, it is particularly important to test the accuracy of its service beam pointing.

The number of beams that each antenna array can split into has become an important indicator of the coverage performance of the Massive MIMO network. The throughput that users can achieve under the coverage of these beams also needs to be part of the evaluation.

Figure 4. Beamforming characteristics

4. Summary

As networks continue to evolve, antennas and RF modules will be deeply integrated, and Massive MIMO active antennas will be the mainstream of future antenna development. Integrated testing and air interface testing may become the evolution direction of future testing.

Compared with traditional antenna and RF testing methods, test indicators and evaluation systems, test principles and methods, test platforms, etc. have all encountered major challenges. These may be unprecedented major innovations in the antenna and feeder networks of mobile communication systems, which urgently need our exploration.