Over-the-Air Performance Estimation of Wireless Device Antennas

Sammanfattning: This thesis work presents studies on the available measurement methods for estimating the over-the-air (OTA) performance of wireless device antennas. This thesis is divided into two parts. The first part is focused on studies of single-input single-output (SISO) OTA measurement methods and the second part is focused on studies of multiple-input multiple-output (MIMO) OTA measurement methods.In the first part, the three methods for estimating the up-link radio performance of mobile phone antennas are evaluated, including the “EMSCAN Lab Express” planar near-field system, the “Telia Scattered Field Measurement method (TSFM),” and the “Bluetest” reverberation chamber. These methods are compared to the standard reference anechoic chamber methods.Each method is experimentally evaluated by measuring the total radiated power from a number of commercially available mobile phones in addition to comparing the measurements to the results from a standard reference system. The comparison is performed statistically using simple regression analysis. For the reverberation chamber (RC), applying simple physical and statistical modeling extends the analysis. In particular, a maximum likelihood (ML) estimator for the Rician K−factor is derived, followed by both statistical and experimental evaluations.Based on the results of the first part of this thesis, it can be concluded that the EMSCAN Lab Express planar near-field method overestimates the uplink performance of mobile phones, consequently introducing an error. The up-link performance of mobile phones estimated using the TSFM method agrees with the reference method but suffers from problems such as extensive measurement time and poor repeatability. The Bluetest RC method estimates the uplink performance of mobile phones well. However, the Rician K−factor computed from the ML estimator is found to be greater than zero, indicating an inadequacy in the propagation environment inside the RC.In the second part, the anechoic chamber-multipath simulator (MPS) and the RC-very large MIMO (VLM) OTA testing methods are studied.In the MPS method, the impact of scattering within the MPS array antennas on the Rician fading statistics is addressed by performing simulations and measurements based on the proposed MPS scattering model and the derived far-field criteria. The results of the investigation suggest that to minimize the impact of scattering, a large MPS system must be designed, although it will result in higher costs. In turn, the costs can be lowered by miniaturizing the MPS system, which can be achieved by following the derived far-field (Fraunhofer distance) criteria. Thus, miniaturization will also allow the measurement uncertainties to be restricted at an acceptable level.In the VLM OTA testing method, the study is based on very large virtual uniform linear array measurements and focuses on the multidimensional characterization of the channel transfer function and the channel impulse response of the reverberation chamber. The results of this study suggest that the reverberation chamber can be used to emulate the more realistic channels for testing the VLM antenna systems.