Algebraic space-time codes for multi-antenna channels and distributed codes for cooperative wireless communication.

摘要

In recent years, multiple-input multiple-output (MIMO) systems have been shown to be significantly beneficial for wireless communication. This is due to an increase in reliability provided by diversity in the system and also an increase in rate provided by the ability of the system to support multiple independent data streams. However, to realize the advantage of MIMO systems in practice, we are faced with the problem of designing high-performance scalable space-time coding schemes with efficient encoding and decoding algorithms. Additionally, hardware requirements may further impose a limitation on the peak-to-mean envelope power ratio (PMEPR) for each antenna. In the first part of this thesis, several new space-time codes are designed under these practical constraints and are derived from algebraic and analytic methods. These space-time codes are delay-optimal, utilize all the available degrees of freedom in the channel, and are based on quadrature amplitude modulation (QAM) symbols. This includes a new two transmit antenna space-time code that exhibits the optimal coding gain among a class of linear codes. A new class of full diversity space-time codes is shown to exhibit low, and in some cases optimal, values of PMEPR. In the absence of channel state information at the receiver, practical space-time coding schemes are presented under the framework of "training codes".; With smaller mobile devices, it may be difficult to implement multiple and sufficiently separated transmit antennas. In this case, one can devise schemes that allow multiple single antenna users separated geographically to cooperate with each other in order to communicate to a common destination. This can potentially lead to a new form of diversity known as cooperative diversity that must be exploited by smart coding schemes. In the second part of this thesis, new practical cooperative coding strategies are presented that are again based on the QAM alphabet. It is shown that the practical coding schemes described here can, in fact, achieve the ultimate asymptotic performance which corresponds to the case of all users coming together to behave as a single user with multiple antennas.