This paper describes a layered space-time (LST) approach for multiple-input multiple-output (MIMO) channels where list decisions of each data substream are used to subtract interference when detecting the next data substream. We study this approach in the context of the IEEE 802.11 standard using orthogonal frequency-division multiplexing (OFDM) and bit-interleaved coded modulation of each subcarrier. We show that ideal list LST with increased list size can potentially approach the MIMO Shannon bound, but performance of real implementation based on a reduced-complexity demapper is limited by competing Euclidean distances common in any reduced-state maximum likelihood detection.

Multicarrier signals are known to suffer from a high peak-to-average power ratio, caused by the addition of a large number of independently modulated subcarriers in parallel at the transmitter. When subjected to a peak-limiting channel, such as a nonlinear power amplifier, these signals may undergo significant spectral distortion, leading to both in-band and out-of-band interference, and an associated degradation in system performance. This paper characterizes the distortion caused by the clipping of multicarrier signals in a peak-limiting (nonlinear) channel. Rather than modeling the effects of distortion as additive noise, as is widespread in the literature, we identify clipping as a rare event and focus on evaluating system performance based on the conditional probability of bit error given the occurrence of such an event. Our analysis is based on the asymptotic properties of the large excursions of a stationary Gaussian process, and offers important insights into both the true nature of clipping distortion, as well as the consequent design of schemes to alleviate this problem.