目前通信系統之設計大部分都集中在低傳輸速率系統之開發，然而，下一代通信系統於網際網路及多媒體上之運用，對於高傳輸率及高機動性的要求以非常快之速度成長，為解決包含時間及頻率選擇性衰減通道(selective fading channel)、功率及頻寬限制所引起效能限制，空時碼(space-time coding)是目前最有效克服衰減之技術及提供高可靠之傳輸。 本論文以三個章節分別探討有關多天線系統之議題，包含非同調(noncoherent)空時碼用於準穩態衰減通道（quasistatic fading channel）之架構、籬柵編碼非同調空時調變（TC-NSTM）及串接空時碼用於不同衰減通道之架構。本論文第三章主要說明在傳送天線及接收天線都不需正確知道通道狀態資訊（CSI）之非同調系統，我們主要提出一種單式空時調變（USTM）以降低二維星座圖訊號點數及保證所尋找出的編碼均為全分集（full diversity），並運用多階層解碼方法以降低解碼複雜度。 第四章提出三種運用於籬柵編碼非同調空時調變之方法，第一種方法為用於區塊衰減通道(block fading channel)環境下之新籬柵編碼單式空時調變(TC-USTM)，此方法使用更多的未編碼單式空時調變信號來挑戰傳統利用二倍未編碼單式空時調變信號之籬柵編碼單式空時調變。第二種方法是利用以訓練碼為基礎(training-based)之籬柵編碼非同調空時調變，運用訓練碼估測通道狀態資訊，並將同一分支訊號利用交錯器（interleaver）將信號打散於不同衰減區塊上，以增加系統分集數。第三種方式是用於準穩態衰減通道之非同調序列偵測演算法，而這些演算法是利用一個滑動視窗(slide window)之維特比演算法(Viterbi algorithm)來完成，這個滑動視窗涵蓋幾個籬柵分支信號。 第五章為提出串接空時區塊碼(STBC)與使用一個延遲處理器（delay processor）之籬柵編碼調變(TCM)架構，且運用於多輸入/多輸出(MIMO)系統及多輸入/多輸出正交分頻多工(OFDM)系統。此架構不僅可獲得較大最小平方歐氏距離(squared Euclidean distance)且可獲得較大最小漢明距離(Hamming distance)，因此，可得到較佳系統性能。本章除提出一種改善式方法，此方法是於空時區塊碼與籬柵編碼調變中間插入一個特殊區塊交錯器(interleaver)，並且提出兩種解碼方法，包含部分硬式決定回授之滑動視窗式之遞迴解碼(SW-IDPHF)及軟式決定回授之滑動視窗式之遞迴解碼(SW-IDSF)。不管是利用那種解碼方式，我們所提出的解碼方法其解碼收斂速度較快且終極性能均比利用傳統解碼方式之位元交錯編碼調變(BICM)為佳。 The main goal of the next generation wireless communication systems provides high data rates and high mobile service for the Internet and multimedia applications. In order to deal with performance-limiting challenges that include time and frequency selective fading channels, power and bandwidth constraints, space-time coding has been proposed as one of the most effective techniques to combat fading and to provide high reliable transmission. This thesis has three chapters to address three important issues, respectively, relative to multiple antennas systems: noncoherent space-time modulation schemes for the quasistatic fading channel, trellis coded noncoherent space-time modulation schemes for various fading channels and concatenated space-time coding with a delay processor for frequency flat/selective fading channels. In Chapter 3, we deal with a noncoherent system in which the transmit and the receive antennas do not know the channel state information (CSI). We consider the design of unitary spacetime modulation (USTM) reduces the number of signal points in the two-dimensional constellation and guarantees full diversity for searched codes as well as reduces the decoding complexity with multistage decoding for large constellation size. Moreover, in the scheme using multistage decoding with power allocation based on USTM, training code and differential space-time block coding (DSTBC) can achieve better error performance than those codes using conventional decoding. In Chapter 4, we propose three schemes for trellis coded noncoherent space-time modulation (TC-NSTM). In the first scheme, based on systematic method for constructing USTM, we propose a new trellis coded unitary space-time modulation (TC-USTM) scheme for the block fading channel; the scheme uses more signals of USTM compared with the existing TC-USTM schemes expanding the signal set of uncoded USTM by a factor of two and can obtain larger distance than existing schemes. In the second scheme, based on training code, a new TC-NSTM scheme is investigated for the block fading channel also. We use training code instead of USTM and insert a proper interleaving to increase temporal diversity. In the last scheme, we generalize the noncohernt sequence detection algorithms to detect TC-USTM over the quasistatic fading channel. The detection using the Viterbi algorithm with a slide window where each window covers several branches of the trellis is proposed. The searched codes for this scheme can achieve better performance than the USTM. In Chapter 5, we propose concatenated space-time block coding (STBC) with trellis coded modulation (TCM) using a delay processor for multiple-input multipleoutput (MIMO) andMIMO-OFDM (orthogonal frequency-division multiplexing) scheme. The scheme can achieve not only minimum squared Euclidean distances but also large minimum Hamming distances and hence can achieve good error performance for both MIMO and MIMO-OFDM systems under various fading channels. The slidingwindow-type iterative decoding using partial hard-decision feedback (SW-IDPHF) and sliding-windows-type iterative decoding using soft-decision feedback (SW-IDSF) are also proposed. The codes can converge fast and get better performance as compared to the case of bit-interleaved coded modulation (BICM) using conventional decoding.
