In this letter, a performance analysis of a dualbranch switched diversity system operating on statistically independent and identically distributed Nakagami-m flat-fading channels is presented. An adaptive coded modulation (ACM) scheme is employed to increase the spectral efficiency of the system. The ACM scheme consists of a set of multidimensional trellis codes originally designed for additive white Gaussian noise channels, where the codes are based on quadrature amplitude modulation (QAM) signal constellations of varying size. The performance is evaluated by assuming perfect channel knowledge at both transmitter and receiver and instantaneous feedback of channel state information, conveyed from the receiver to the transmitter on an error-free feedback channel. The optimal switching threshold of the switched diversity combiner, maximizing the average spectral efficiency, is identified for spatially uncorrelated antenna branches.

In this paper we investigate the performance of an adaptive coded modulation (ACM) system in single-input multiple-output Rayleigh fading channels with spatial correlation. The considered ACM system has a set of multi-dimensional trellis codes from which it can adaptively select the code best suited to the channel quality. The channel quality is tracked at the receiver by a predictor using a pilot-symbol-assisted modulation scheme. We optimize both the pilot spacing, and the power allocation to pilot and data symbols. The channel predictor is made optimal in the maximum a posteriori sense and the correlation to other subchannels is taken into account. A numerical example is given for flat Rayleigh fading subchannels with Jakes correlation profile.

In this paper, a performance analysis of a dual- branch switched diversity system operating on identically distributed Nakagami-m fading channels is presented. An adaptive coded modulation (ACM) scheme is employed to increase the spectral efficiency of the system. The ACM scheme consists of a set of multidimensional trellis codes originally designed for additive white Gaussian noise channels, where the codes are based on quadrature amplitude modulation (QAM) signal constellations of varying size. The performance is evaluated by assuming perfect channel knowledge at the receiver and instantaneous feedback of channel state information, conveyed from the receiver to the transmitter on a zero-error feedback channel. Both spatially uncorrelated and correlated antenna branches are considered. The optimal switching threshold in terms of maximizing the average spectral efficiency is identified in the case of uncorrelated antenna branches.