Among various diversity techniques to combat fading in wireless channels, spatial diversity through MIMO coding scheme is an effective way to increase link capacity and system reliability without sacrificing bandwidth efficiency. Recently, cooperative diversity has been introduced as an efficient alternative to improve system performance without the requirement of additional antennas. However, most of existing works on MIMO and cooperative communications are based on an assumption that the destination has perfect knowledge of channel state information of all transmission links and hence introduces high complexity to the receiver.

To overcome such problems, this thesis proposes differential modulation schemes for space-time coded MIMO and cooperative communications. By exploiting spatial/cooperative diversity without the requirement of channel state information, the proposed schemes provide an excellent tradeoff between receiver complexity and system performance. First, a matrix rotation based signal design for differential space-time modulation is investigated to minimize the union bound on block error probability. Next, a robust differential scheme for MIMO-OFDM systems is proposed by which the signal transmission of each differentially encoded signal is completed within one OFDM block rather than multiple blocks as in existing works. Then, a differential scheme for UWB systems employing MIMO multiband OFDM is proposed to explore all available diversities by jointly encoding across spatial, temporal, and frequency domains. To exploit cooperative diversity, an amplify-and-forward differential cooperative scheme and a threshold-based decode-and-forward differential cooperative scheme are proposed. The proposed differential cooperative schemes are first considered in a two-node cooperation system, and the proposed works are extended to a general multi-node scenario. Finally, a general framework to improve lifetime of battery-operated devices by exploiting cooperative diversity is proposed such that the device lifetime can be greatly improved by efficiently taking advantages of both different locations and energy levels among distributed nodes in wireless networks.