Contents I. Introduction
Future wireless systems are likely to be equipped with multiple antennas. Hence, adaptive-transmission techniques already implemented in single-input single-output (SISO) channels have to be extended to encompass the features of the multiple-input multiple-output (MIMO) fading channel. In an MIMO system, the use of multiple antennas adheres to one of two distinct approaches that seek to improve either the diversity gain or the information rate of the system. These two approaches are commonly referred to as MIMO diversity and spatial multiplexing, respectively [1]. Herein, we consider the former approach and investigate the Shannon capacity of MIMO Rayleigh fading channels using orthogonal space-time block coding (STBC). The Shannon capacity of a channel characterizes its maximum achievable rate, given no delay or complexity constraints for an arbitrarily small bit error rate. In practice, different power- and rate-allocation policies that allow for different compromises between the achievable capacity and the corresponding implementation complexity can be performed when a channel state information (CSI), consisting of the signal-to-noise ratio (SNR) as estimated by the receiver, can be made available to the transmitter. In this paper, we derive closed-form expressions for the Shannon capacity of MIMO Rayleigh fading channels using STBC for three adaptive-transmission policies: 1) optimal power and rate adaptation (opra); 2) optimal rate adaptation (ora), given constant transmit power, and 3) channel inversion with fixed rate adaptation (cifr).
