多輸入多輸出(multiple-input multiple-output, MIMO)無線通訊系統結合中繼網路可以改善訊號之覆蓋範圍並且利用空間分集(spatial diversity)增加傳輸可靠度。然而,相對的也提升射頻電路的個數與系統複雜度,使得成本也提高,為了減少電路成本並允許犧牲少許錯誤率,必須從眾多天線中有效的選擇最適合的天線子集來傳送訊號。因此本論文以相關性通道中放大轉發方式(amplify-and-forward, AF) MIMO多中繼系統為架構,針對晶格縮減(lattice reduction, LR)中常見的CLLL (Complex Lenstra–Lenstra–Lovász)演算法提出縮小(size reduction, SR)搭配晶格縮減的SR-LR演算法用以降低複雜度並且維持近似的位元錯誤率,並將其應用至相關性通道中多中繼節點的天線選擇。最後藉由電腦模擬相關應用的演算法可以得知在相同的位元錯誤率下能加以降低複雜度。
In this thesis, an amplify-and-forward (AF) multiple-input multiple-output (MIMO) multiple-relay system is considered to improve the coverage of signal and provide better reliability with the benefit of signal transmitting spatial diversity. However, the multiple radio frequency (RF) chains associated with multiple antennas are known to be quite costly. In order to reduce the cost of RF chains, a size reduced and lattice reduction (SR-LR) algorithm is proposed. The proposed SR-LR algorithm with the use of size reduction reduces the complexity of the CLLL (complex Lenstra–Lenstra–Lovász) with lattice reduction. The proposed SR-LR algorithm is applied to multiple-relay antenna-pair selection in case of correlated channels in both hops. Simulation results demonstrate that the proposed SR-LR algorithm is able to maintain the BER performance at a much lower computational complexity.