Satoru Fukumoto

Field Experimental Results on E-UTRA Downlink Throughput Using Polarization Diversity Antennas

This paper investigates the effect of the antenna (polarization diversity (PD) or space diversity (SD)) configuration at evolved Node B and antenna angle of user equipment (UE) on the throughput performance of multiple input multiple output (MIMO) in Evolved UMTS Terrestrial Radio Access (E-UTRA) field experimental system. The results show that the throughput performance as a function of the measured signal-to-interference plus noise power ratio (SINR) per antenna for the PD configuration is degraded by approximately 1 dB and 3-6 dB in a 2times2 MIMO case for non-line of sight (NLOS) and line of sight (LOS) environments, respectively, and 1-2 dB in a 4times4 MIMO case for a NLOS environment, when the angle of the receiver antenna is 0 degrees due to variation in the average received power gap between vertically-polarized and horizontally-polarized waves. On the other hand, when the receiver antenna angles are 45 or 90 degrees, the throughput performance for the PD configuration is almost the same as those for the NLOS environment and is improved compared to those for the LOS environment, respectively, due to a smaller average received power gap and lower fading correlation. It is also shown that the cumulative distribution function (CDF) of the throughput performance when using the PD configuration is nearly identical to those for the 2times2 and 4times4 MIMO cases irrespective of the UE antenna angle for the NLOS environment. The CDF in this case is also better than those for the SD configuration when the UE antenna angle is 90 degrees for the NLOS environment and 45 and 90 degrees for the LOS environment.

Field Experimental Results of E-UTRA Downlink with Precoding and Non-Precoding MIMO

This paper investigates the effect of precoding in multiple input multiple output (MIMO) transmission from the viewpoint of transmitter antenna configurations, i.e., space diversity (SD) or polarization diversity (PD), in field experiments using the Evolved UMTS Terrestrial Radio Access (E-UTRA) experimental system. The field experimental results show that there is no performance gap between with and without precoding irrespective of the antenna configuration in a non-line of sight (NLOS) environment when a common precoding vector is applied over the entire signal bandwidth of 18 MHz and the maximum likelihood detection with QR-decomposition and M-algorithm (QRM-MLD) based algorithm is applied at the receiver. In a high correlation channel as observed in a line of sight (LOS) environment, there is also no performance gap between with and without precoding. However, in the high received SINR region, the throughput performance of the precoding with the PD configuration is worse than that without precoding since decoding errors occurred in both codewords, which are mapped onto two different transmitters, due to a significantly high received power gap between vertically and horizontally-polarized antennas.

Field experiment results using 2×2 and 4×4 MIMO precoding in E-UTRA downlink

We present field experiment results on multiple input multiple output (MIMO) precoding using orthogonal frequency division multiple access (OFDMA) in the E-UTRA downlink. We measured the throughput performance of 4×4 MIMO precoding with 1- to 4-stream transmission. Time variation characteristics of throughput, SINR, and antenna correlations are compared to evaluate the selection criteria for the number of streams. The throughput performance of 2×2 MIMO precoding are also evaluated and compared with the results of the 4×4 MIMO case. Based on the evaluation of the field experiment results, we verify that the throughput is maximized using 3- or 4-stream transmission. Furthermore, we verify that 2-stream transmission is also effective for 4×4 MIMO in an actual propagation environment, since high antenna correlation sometimes happens on which the throughput performance of 3- or 4-stream transmission is severely degraded. Evaluation results also show that the estimated gain of rank adaptation is about 11.6% for 4×4 MIMO and about 6% for 2×2 MIMO.