Abstract
The main task of the next generation of wireless communication systems is aimed at providing high data rates and high transmission reliability. However, in practical wireless communication systems, the signal transmission experiences channel links with rapid fading coefficients, and it is usually difficult to precisely track the channel variations. The accuracy of the channel estimation may strongly influence the transmission reliability of the entire communication system. Consequently, the designs for noncoherent coded systems, where neither the transmitter nor the receiver requires the information of the channel coefficients in advance, are desirable. In this thesis, three topics are addressed, which are respectively presented from Chapter 2 to Chapter 4. In Chapter 2, the focus is on a single-input single-output (SISO) noncoherent communication system. For uncoded schemes, the pair-wise error probability (PEP) and its high-SNR approximation for the signal vectors with non-uniform power values under noncoherent block fading channels are derived. For coded schemes, a method that approaches the capacity limit for the noncoherent block fading channel using a codeword-interleaving strategy is proposed. The labelling/interleaving rules are also investigated. It is shown that with the assistance of a specially-designed iterative receiving algorithm, the overall coded scheme achieves a near-capacity performance, which is about 1.6 dB away from the capacity limit with constrained input signals. In contrast, in Chapter 3 and Chapter 4, the focus is on modulation and detection strategies for a multiple-input multiple-output (MIMO) noncoherent communication system. In Chapter 3, a noncoherent MAP (maximum {\it a posteriori}) demodulator with very low complexity is proposed for orthogonally designed USTM (unitary space-time modulation) schemes. In addition, it is shown that the detected soft values are independent of the extrinsic information fed back from the channel decoder, when Gray-mapped QPSK signal is adopted. As a result, iterative detection and decoding between the noncoherent demodulator and the channel decoder can be avoided without any degradation in performance. Compared to schemes that use a exhaustive-search-based noncoherent MAP demodulator, the proposed scheme achieves the same error performance, and has a much lower detection complexity. In Chapter 4, a novel turbo coded space-time modulation scheme for noncoherent block fading channels is presented. The coded bits are divided into two parts, which are then respectively transmitted using USTM and spatial multiplexing (SM). Since this approach effectively increases the cardinality of the set of possible transmit (space-time) signal matrices, a large rate gain can be obtained without increasing the modulation order. At the receiver, an iterative detection-decoding algorithm is performed cooperatively among the turbo decoder, the coherent demodulator for the SM, and the noncoherent demodulator for the USTM. Compared to the previous work in the literature, the proposed scheme can provide significant advantages in both error performance and complexity. The overall designed coded scheme achieves a near-capacity performance, which is 0.7 dB away from the constrained input capacity limit.