於多天線空中測試系統中結合天線選擇之三維空間通道仿真

Abstract

於本論文中，吾人首先在三維測試系統中利用空間相關性 (spatial correlation)並考慮三維功率頻譜之形狀提出通道仿真之最佳化問題，並解出每根天線的最佳能量權重。由於使用傳統方法實現通道仿真時，於某些角度會產生過大的誤差，吾人提出極小極大法 (min-max) 去抑制潛在性的仿真誤差。然而，在三維測試系統中，設備成本為重要考量，吾人提出天線選擇法降低選用天線數，並維持仿真之準確度。為了更進一步提升測試真實性，吾人提出在真實環境中利用最小變異無失真響應 (minimum variance distortionless response, MVDR) 技術取得真實環境中之角度資訊並在無反射電波暗室 (anechoic chamber) 中重現通道特性。經由模擬驗證，吾人提出的能量權重計算方法相較於傳統方法，可以達到較佳的精準度，並能根據真實環境中估計到的角度資訊計算出準確的能量權重，藉此在無反射電波暗箱中重建通道特性。

In this thesis, we formulate a channel emulation problem with practical constraints in a 3D MIMO OTA testing system and solve for the optimal power weights for each probe antenna. When using conventional algorithms to calculate the optimal power weights, the reconstruction errors may be large in some directions. We propose a min-max algorithm to suppress the maximum error. Moreover, to reduce the setup cost, we propose a probe antenna selection algorithm to reduce the number of active probe antennas while keeping the accuracy of the testing. In addition, we use the minimum variance distortionless response (MVDR) beamformer to estimate the angle information in real environments and regenerate the channel characteristics in the anechoic chamber. Finally, simulation results show that the proposed algorithm presents a better accuracy compared with the conventional algorithms and can determin the optimal power weights accurately based on the estimated angle information to reconstruct the channel characteristics in the chamber.