用於多輸入多輸出正交分頻多工系統之IQ不平衡與相位雜訊補償技術

Abstract

在目前行動通訊系統講求更高傳輸速率以及更低消耗功率情形下，多輸入多輸出正交分頻多工系統結合直接降頻接收器逐漸被廣為採用。然而，但在射頻端會產生某些不理想之效應，例如IQ不平衡以及相位雜訊，這兩種效應皆會對通訊系統產生不同的影響，使得通訊系統之性能下降。在本論文中，吾人提出在多輸入多輸出正交分頻多工系統下補償IQ不平衡以及相位雜訊之方法，吾人利用系統前置碼與領航符元同時估計IQ不平衡、相位雜訊以及通道效應，並利用所估計之值與期望值最大化演算法補償IQ不平衡與相位雜訊，藉由期望值最大化演算的疊代，對通道響應與相位雜訊更進一步估計，其估計值再回饋至IQ不平衡補償器中重新再做補償。吾人藉由模擬結果顯示利用疊代以及回饋之方法可將非理想效應消除。當非理想效應獲得消除，射頻端的電路設計及能大為簡化。

Multiple-input multiple-output orthogonal frequency-division multiplexing (MIMO-OFDM) systems are often realized with direct-conversion architectures for higher throughput and lower power consumption. However, such systems suffer in radio-frequency (RF) impairments such as in-phase/quadrature-phase (IQ) imbalance (IQI) and phase noise (PN) which significantly degrade the performance of communication systems. In this thesis, we propose an iterative method to compensate IQI and PN in MIMO-OFDM systems. We exploit preamble and pilot symbols to estimate the channel, IQI and PN effects, and utilize the estimated value and expectation-maximization (EM) algorithm for eliminating the impairments. The channel effect and PN can be estimated further with the iterative scheme and then fed back to the compensator to compensate the IQI again. We show by simulations that improvement in performance is possible with the proposed iterative method. When the effects of RF impairments are alleviated, the design of RF and analog circuitry can be greatly simplified.