Evaluation of MIMO OTA parameters for LTE using a Mode-Stirred Reverberation Chamber

Transcription

1 1 EUROPEAN COOPERATION COST 2100 TD(10)12073 IN THE FIELD OF SCIENTIFIC Bologna, Italy AND TECHNICAL RESEARCH 2010/Nov/23-25 EURO-COST SOURCE: EMITE Ing 1, AT4 Wireless 2, Technical University of Cartagena 3 Spain Evaluation of MIMO OTA parameters for LTE using a Mode-Stirred Reverberation Chamber Juan D. Sánchez-Heredia 1, Juan F. Valenzuela-Valdés 1, Juan P. Hidalgo 2, Alejandro Torrecilla 2, Sergio Lobato 2, David A. Sánchez-Hernández 3 1 EMITE Ing 2 AT4 Wireless Edificio CEEIM Campus Espinardo Parque Tecnológico de Andalucía C/ Severo Ochoa, 21. Parcela I+D 6B E Murcia Málaga SPAIN Phone: SPAIN Phone: Fax: Fax : support@emite-ingenieria.es jphidalgo{atorrecilla}@at4wireless.com 3 Technical University of Cartagena Antiguo Cuartel de Antigones Plaza del Hospital, 1 E Cartagena SPAIN Phone: Fax : david.sanchez@upct.es

2 Evaluation of MIMO OTA parameters for LTE using a Mode-Stirred Reverberation Chamber Juan D. Sánchez-Heredia 1, Juan F. Valenzuela-Valdés 1, Juan P. Hidalgo 2, Alejandro Torrecilla 2, Sergio Lobato 2 and David A. Sánchez-Hernández 3 1. Abstract 1 EMITE Ing Edificio CEEIM. Campus Espinardo E Murcia, SPAIN support@emite-ing.com 2 AT4 Wireless Parque Tecnológico de Andalucía C/ Severo Ochoa, 21. Parcela I+D 6B Málaga, SPAIN jphidalgo{atorrecilla}@at4wireless.com 3 Technical University of Cartagena Antiguo Cuartel de Antigones. Plaza del Hospital, 1 E Cartagena, SPAIN david.sanchez@upct.es Several methodologies have been proposed for standardized MIMO OTA testing for 4G systems, which includes the reverberation chamber, a two-stage method and a multiple test probe method. The different methodologies differ in terms of complexity and cost-effectiveness. In order to adequately evaluate the overall MIMO performance of mobile terminals equipped with multi-antennas for the receive diversity and MIMO transmission, several figures of merit have been proposed. In order to verify the validity of the proposed MIMO OTA testing methodologies, a round robin campaign for LTE devices has been scheduled by 3GPP/CTIA. In this contribution, we present the report of the MIMO OTA test results for LTE devices using a multicavity mode-stirred reverberation chamber for the first time. The E400 MIMO Analyzer modestirred reverberation chamber by EMITE Ing, an LTE-enabled E2010 Broadband Wireless Test Set Base Station Emulator by AT4 Wireless and a Samsung GT-B3710 LTE USB Dongle DUT have been employed. Results show that it is possible to perform standardized MIMO OTA tests with mode-stirred reverberation chambers with same accuracy than those reported for other methodologies. 2. Introduction A Study Item of the MIMO OTA testing for multiple antenna terminals was agreed at the 3GPP standardization committee. The main purpose of the SI is to gather sufficient data so as to be able to establish commonly acceptable testing methodologies for an adequate evaluation of the overall MIMO performance of wireless communication terminals with multiple antennas. Ultimately, the testing methodologies have to be able to differentiate a good from a bad MIMO device, and a set of different figures of merit (FoMs) has been agreed for that purpose. In order to gather and compare testing results, an LTE MIMO OTA round robin test campaign has been jointly organized by 3GPP, CTIA and COST2100 [1]. A the 3GPP RAN4 AdHoc#4 meeting in Xian, China, October 2010, EMITE Ing was set to be the first company to start the LTE MIMO OTA Round Robin Campaign for 3GPP/CTIA/COST2100. In the 3GPP RAN4 #57 Meeting in Jacksonville, USA, a preliminary list of companies was made, and it is reproduced in Table 1, and other interested proponents were encouraged to commit to the test plan by formally asking 3GPP to participate on the LTE MIMO OTA Round Robin Test Plan. The list of participants is shown below: 1

3 Member Country DUT Comments EMITE Ing. Spain To be Provided by Vodafone Huawei China To be Provided by Vodafone Spirent communications USA To be Provided by Vodafone R&S Germany To be Provided by Vodafone Agilent China To be Provided by Vodafone Azimuth Systems USA To be Provided by Vodafone TMC China To be Provided by Vodafone Table 1: 3GPP/CTIA/COST2100 LTE MIMO OTA Round-robin Measurement Campaign The objective of this contribution is to present the test conditions and some throughput results used by EMITE Ing for 3GPP/CTIA/COST2100 LTE MIMO Round Robin Measurement campaign using a modestirred reverberation chamber. To our knowledge this is the first time these results (LTE MIMO OTA) are reported using a mode-stirred reverberation chamber. 3. Test conditions As a part of 3GPP/CTIA/COST 2100 LTE MIMO OTA Round Robin Measurement Campaign, EMITE Ing has run diverse tests to an LTE-enabled DUT. The tested DUT was provided by Vodafone, and it is a GT-B3710 LTE USB Dongle by Samsung, which was used attached to a HP-Compaq 6720s laptop. Following the test plan instructions [2], the lid of the laptop was closed. The center of the rotation was the tree dimensional geometric centre of the laptop. The DUT was connected to the laptop to the single USB-port on the left side. For the testing of the DUT, the laptop was used as a host device with its internal radios turned off. The laptop was always powered by battery in all tests, and power management settings used were according to the test plan: - Turn off monitor never - Turn off hard disks never - System standby - never - WLAN-off - Bluetooth radio-off The DUT was tested using the E400 MIMO Analyzer mode-stirred reverberation chamber in connection to an AT4 Wireless E2010 Wireless Broadband Test Set LTE Base Station Emulator. The E400 MIMO Analyzer is a mode-stirred reverberation chamber equipped with 9 fixed measurements antennas to enable MIMO measurements. When the E400 mode-stirred reverberation chamber was connected to the AT4 E2010 BSE, it was tuned to the NIST Indoor-Urban channel model which is based on real outdoor-to-indoor channel measurements in urban environments [3-4]. This is like the exponential decay specified in the Test Plan. This is done by careful tuning of the chamber RMS delay spread (RMS DS) through a very accurate 2

4 fitting of the power delay profile (PDP) [5-6]. Testing was performed using step-wise stirring, wherein the throughput was sampled at each fixed stirrer position to avoid any Doppler shift. A set of three repetitions was performed for each measurement, and averaged values over these three repetitions are provided as final values. For other channel model emulation, either the AT4 E2010 LTE BSE was employed with its embedded extended Pedestrian A (EPA) channel model emulation. Both open-loop and closed-loop spatial multiplexing transmission modes were tested. Measurements in closed-loop mode were done using fixed values of rank and precoding. Other test conditions are summarized in Table 1. Table 1. Testing conditions of the LTE MIMO OTA testing. DUT Samsung GT-B3710 LTE device Category 3 Channel bandwidth Channel power (Signal level at DUT) Figure of merit Modulation and Coding Scheme (MCS) 10, 20 MHz -120 to -60 dbm Throughput (FRC) 13 (16QAM) and 26 (64QAM) Number of RB 50, 100 Spatial channel models f d NIST / EPA 5 Hz The BSE parameters were set according to the type of DUT to be tested. For the purpose of the LTE MIMO OTA Round Robin, only the spatial multiplexing modes were used (open-loop and closed-loop spatial multiplexing). The employed DUT Category, 3, and their mapping to the Fixed Reference Channels (FRCs) are given in Table 2. The AT4 E2010 parameter settings were those in Table 3, as specified in the test plan. UE category Category 3 Data rate (DL / UL) (Mbps) 100 Mbps / 50 Mbps Table 2. Mapping of LTE DUT Categories into FRC. DL UL Max. num.of Max. num. of Total Max. Max. Max. num. of Support DL-SCH TB DL-SCH num. of num. of num. of UL-SCH bits for bits bits soft spatial UL-SCH per TB 64QAM per TTI per TB channel layers TB bits per TTI per TTI bits per TTI No FRC FRC1 Table 3. Main BSE parameter settings. Parameter Unit Value FRC1 Channel bandwidth MHz Modulation and coding scheme (MCS) 16QAM 16QAM Target Coding Rate 1/2 1/2 Allocated resource blocks RB Subcarriers per resource block Allocated subframes per Radio Frame Number of HARQ Processes 8 8 Maximum number of HARQ transmissions 1 1 Information Bit Payload per Sub-Frame For Sub-Frames 1,2,3,4,6,7,8,9 Bits For Sub-Frame 5 Bits n/a n/a For Sub-Frame 0 Bits Transport block CRC Bits Number of Code Blocks per Sub-Frame 3

5 For Sub-Frames 1,2,3,4,6,7,8,9 Bits 3 5 For Sub-Frame 5 Bits n/a n/a For Sub-Frame 0 Bits 3 5 Binary Channel Bits Per Sub-Frame For Sub-Frames 1,2,3,4,6,7,8,9 Bits For Sub-Frame 5 Bits n/a n/a For Sub-Frame 0 Bits Maximum throughput averaged over 1 Mbps 11, ,9100 frame (per data stream) Noc dbm / khz Precoding granularity PRB PMI delay ms n/a n/a Reporting Interval ms n/a n/a Reporting Mode n/a n/a PDCCH Number of OFDM symbols for PDCCH 2 2 Aggregation level 8 8 DCI format 2A 2A Cell ID 0 0 All tested radio environments were calibrated as specified in the test plan [2]. The evaluated Figure of Merit (FoM) was MIMO Throughput (FRC Category I). 4. Test results Figure 1 shows the Throughput test results of LTE MIMO OTA for the evaluated DUT under a NIST Indoor-Urban and an EPA fading channel model using a 10 MHz channel bandwidth and for open-loop spatial multiplexing transmission mode. Figure 2 shows the Throughput test results of LTE MIMO OTA for the evaluated DUT under a NIST Indoor-Urban fading channel model using a 10 MHz channel bandwidth and for closed-loop spatial multiplexing transmission mode using 2 different Modulation Coding Scheme (MCS), in order to compare 16QAM and 64QAM modulations. 4

6 Figure 1. Test results of LTE MIMO OTA Throughput for the Samsung GT-B3710 LTE USB Dongle under a NIST Indoor-Urban and an EPA fading channel model and for 10 MHz channel bandwidth and open-loop spatial multiplexing transmission mode. Figure 2. Test results of LTE MIMO OTA Throughput for the Samsung GT-B3710 LTE USB Dongle under a NIST Indoor-Urban fading channel model and for 10 MHz channel bandwidth and closed-loop spatial multiplexing transmission mode, using different MCS 13 (16QAM) and MCS 26 (64QAM). 5

7 Figure 3 shows the Throughput test results of LTE MIMO OTA for the evaluated DUT under a NIST Indoor-Urban fading channel model using both 10 and 20 MHz channel bandwidth and for open-loop spatial multiplexing transmission mode. Figure 3. Test results of LTE MIMO OTA Throughput for the Samsung GT-B3710 LTE USB Dongle under a NIST Indoor-Urban fading channel comparing both 10 and 20 MHz channel bandwidth using open-loop spatial multiplexing transmission mode. It can be confirmed from these figures that the throughput is definitely decreased according to the decrease in the channel power. It is also clear that the channel bandwidth has a clear impact on Throughput, as expected. Clear differences in performance of the analyzed DUT when the channel bandwidth is changed are observed, as expected. Yet, important differences in performance of the analyzed DUT are observed when the fading channel model is changed for each channel bandwidth. 5. Conclusions In this contribution, we have presented some results of the 3GPP/CTIA/COST2100 LTE MIMO OTA Round Robin testing using an EMITE E400 mode-stirred reverberation chamber in connection to an AT4 Wireless Broadband Test Set LTE Base Station Emulator and a Spirent SR5500 Channel Emulator. Results show that the Samsung GT-B3710 LTE USB dongle performs differently to different channel bandwidths, as expected, but that it responds well to the variation of fading channel models. Testing over more LTE-enabled DUTs is needed to help differentiating a good from a bad MIMO device, which is the aim of the Study Item at 3GPP. Results also show that the slope of the Throughput curve is larger than the slopes found previously for HSDPA DUTs, which call for a simple a fast new figure of merit that could use all the generated data in the round robin to come up with the distinction between a bad from a good MIMO device. 6

8 Results also show that the active MIMO OTA testing using mode-stirred reverberation chambers is a good premise to evaluate the real LTE MIMO OTA performance of wireless devices and to differentiate a good from a bad MIMO device. 6. References [1] Vodafone, Test Plan of MIMO/Multiple Receiver Antennas OTA Measurement Campaign. [2] 3GPP RAN4 Tdoc R by Vodafone. Revised LTE MIMO OTA TSG-RAN Working Group (Radio) meeting #57. Jacksonville, Florida, USA, November, [3] RCSG090914, Baseline Criteria for SIMO/MIMO Radiated Performance Testing, AT&T, CTIA Reverberation Chamber Subgroup contribution, September [4] RCSG090913, Outdoor-to-Indoor Channel Measurements and Models, D. Matolak, K. Remley, C. Holloway, CTIA Reverberation Chamber Subgroup contribution, September [5] Delangre, O. et al., Modeling in-vehicle wideband wireless channels using reverberation chamber theory, Proceedings of the IEEE Vehicular Technology Conference, pp , 2007 [6] Delangre, O. et al., Delay spread and coherence bandwidth in reverberation chamber, Electronics Letters, Vol. 44, No. 5, pp , Feb