Modern wireless communication technologies, such as LTE, HSPA + and WiMAX, all use MIMO (Multiple Input/Multiple Output) technology to improve the speed of data transmission. Accordingly, the antenna test system (OTA) also needs to upgrade to the test mode supporting MIMO. This paper presents a simple and economical test method to verify the OTA performance of MIMO equipment, including isotropic sensitivity (EIS) and throughput.
1 MIMOOTA measurement requires new measurement methods.
A high-quality wireless communication device requires an almost omnidirectional antenna, and OTA (antenna performance test) is used to evaluate the performance of the antenna. Nowadays, OTA testing is an important test item in wireless device authentication testing: in the darkroom environment, get the radiation emission power and reception sensitivity of the three-dimensional pattern. At present, the standard OTA test is SISO mode (single input / single output), such as mainstream 2G, 3G and WLAN 802.11a, b, g and other equipment, the main test indicators are TRP (total radiation power) and TIS (total isotropic sensitivity), using the standard CTIA or 3GPP series.
Now, MIMO technology has been widely used in order to improve the network performance of data application. With the use of space division multiplexing (SDM) technology, the channel capacity of spectrum allocation can be significantly improved, but the new transmission technology also brings new testing requirements.
2 *2MIMO has two downstream data streams, so UE (user equipment) needs two receiving antennas. In order to achieve the best performance, we want to minimize the correlation between receiving antennas, so as to facilitate the simultaneous reception of different data streams. In order to evaluate the performance of the whole receiving antenna, it is not enough to evaluate the performance of each receiving antenna separately. The OTA performance test of MIMO equipment needs to be measured under the following working mechanism:
Transmit diversity mode (redundant data flow, improve receiver sensitivity).
* space division multiplexing mode (multichannel data flow, increase throughput).
The modem of LTE usually works indoors. The objective multipath effect makes it possible to achieve multipath reception from various angles. Therefore, in OTA testing, it is also necessary to carry out full three-dimensional analysis and evaluation, that is, the spherical measurement method provides similar conditions to the actual work.
2 efficient MIMOOTA testing method: dual channel test scheme
The OTA testing of SISO in 2G and 3G is mandatory. Usually we hope that the original antenna chamber can directly meet the requirements of SISO and MIMO applications. We also hope that the existing SISOOTA testing system can be easily upgraded to MIMOOTA testing. The dual channel test method of Rhodes and Schwartz can meet this need conveniently.
Dual channel measurement is a direct and effective way to test the OTA performance of MIMO devices. At the same distance from UE, two dual polarization and rotating incident angle test antennas are placed to transmit different MIMO downlink signals. The general characteristics of UE antennas can be obtained by combining various azimuth and polarization modes. As shown in Figure 1, the OTA darkroom contains four angular locators: an angular locator, two test antennas: ANTDL1, ANTDL2, with an angular distribution of 10 degrees (simulated countryside) or 90 degrees (simulated city); and a communication antenna: ANTUL; besides, a radio frequency access board allows five RF channels to be connected to the darkroom. An internal antenna; external equipment includes a base station simulator (R&SCMW500) and a switching matrix (R&SOSP130).
Fig. 1 TS8991 block diagram of MIMO OTA test system supporting dual channel method
The dual channel test method can also be used to verify the smart antenna whose pattern and load impedance can be adaptively changed with the environment. Since no auxiliary RF cable is needed to connect to the port of the test antenna during the test process, the impedance characteristics of the EUT antenna can be ensured to be completely consistent with the actual use.
3 results of conduction and radiation tests
In fact, the receiving antenna module with a certain pattern can be regarded as an additional correlation component, which usually increases the correlation of the input data stream of the receiver, thus affecting the performance of the whole MIMO reception. Therefore, we hope that the ideal MIMO antenna will not have any additional impact on the correlation of data streams.
In order to characterize the overall MIMO performance of the device, it is necessary to simulate different environmental fading scenarios, and simulate different directional characteristics in each scenario. There are different incidence angles and different distribution of polarization scenarios in real environment. To characterize the correlation of antennas, in theory, the radiation characteristics of the receiving antenna can be described completely by choosing two channels to represent different orientations and polarization directions, and selecting appropriate test parameters.
However, the receiving characteristics of the user equipment depend not only on the characteristics of the antenna system, but also on the performance of its own receiver module. Therefore, we propose to divide the whole MIMO test into two separate measurement steps: conduction test and radiation test.
Conduction testing is part of 3GPP compulsory certification, which is the decline test we often talk about. The purpose of this part of the test is to verify the performance of UE receiver in the dynamic change of communication channel parameters.
The main concern of this paper is the radiation test, namely the dual channel OTA test, is used to test the correlation of the UE receiving antenna. Radiometric testing (OTA) is complementary to conduction testing, which greatly reduces the complexity of the test. Compared with a complete multipath propagation scenario system, the test cost is greatly reduced. Base station simulator needs for conduction and radiation measurements.
Figure 2 shows the conduction measurements of 4 different LTE modems. The test uses downlink 2 *2 MIMO open loop space division multiplexing, 16QAM modulation, data throughput and absolute downlink power function curve as shown in the figure. In this case, two RF cables are used to connect the BSE to the UE antenna port, and the results provide the transmission mechanism and the sensitivity performance reference value under the BSE setting. Because there is no coupling between two channels, the maximum spatial polarization is achieved at this time. That is to say, the throughput test of MIMO has the best sensitivity in conduction mode, where the reception performance depends only on the performance of the UE receiver.
Figure 2 conduction test mode, function of data throughput and absolute downlink power.
Seen in Figure 2. UE2 and UE4 have the best receiving sensitivity. The sensitivity of UE1 is about 4dB. UE1, UE2 and UE3 work at frequency band 7 (2.6GHz). UE4 operates at frequency band 20 (800MHz).
In radiation test mode, the transmitting antenna tests the UE at different spatial azimuths, and there are four combinations of polarization modes per azimuth (see Figure 3). The average throughput (RS EPRE) is the linear average of all test results. UE2 uses its built-in antenna; UE2 ExtAnt uses two separate half-wavelength orthogonally polarized external antennas connected to the UE2 antenna port, which provides the best MIMO antenna performance because the external antenna arrangement ensures maximum spatial and polarization dividers.
Fig. 3 radiation test mode, function of data throughput and absolute downlink power.
The performance of UE4's MIMO antenna has greatly reduced its performance. Even though it achieves the best MIMO sensitivity in conduction testing, its average throughput in radiation testing is poor. To achieve half of the maximum throughput, UE4 needs more than 5dB of downlink power than UE2. The measured results are consistent with the theoretical expectations: UE4 operates at a frequency band of 20 (800MHz) and has a wavelength about three times that of band 7. However, UE4 is as compact as UE2, so it is difficult to design the antenna to achieve the same spatial and polarization divider effect as UE2.
In addition to throughput, other OTA parameters can also be obtained by dual channel methods, such as EIS, TIS and so on. The EIS (effective isotropic sensitivity) 3D distribution of MIMO can be automatically generated by R&S AMS32 system software (see Figure 4), where the received power corresponds to a 5% block error rate (BLER).
4 standardization process of MIMO OTA test method
The most relevant standardized entities that are related to the OTA standard are the CTIA and 3GPP RAN4 of the United States. At present, various MIMO OTA test schemes are submitted and discussed. The final determination of the test method depends on the quality of the test results (in other words, whether good design and bad design can be judged), as well as on the complexity of the system, which is closely related to the investment cost of the solution. Besides, testing time is also a consideration. Europe's "COST action 2100" has laid the foundation for MIMO OTA testing. The new "COST action IC1004" was launched in the summer of 2011, which will greatly promote the standardization process.
In order to obtain a complete description of the performance of MIMO OTA, two transmitting antennas are required to test the UE antenna in three-dimensional direction at each incident angle, and a combination of different polarization modes is also required. The combination of conduction test and radiation test can be used to determine whether the problem lies in the antenna design of the wireless equipment or the performance of the receiver itself.
Dual channel test can also be used to test multipath propagation effect (fading). At this point, it is necessary to add a R&S AMU200A baseband signal generator with fading module to the TS8991 test system of MIMO OTA to simulate the downstream data stream generated by R&S CMW500.