In this paper, we analyze the impact of polarization diversity on the capacity of multiple-input multiple-output (MIMO) channels in indoor environments. A channel measurement campaign was conducted at 2.4 GHz to measure the co-polarized and crosspolarized subchannels under line-of-sight (LOS) and non-line-ofsight (NLOS) channel conditions. We analyze the measured data in terms of Ricean K-factor, cross-polar discrimination (XPD) and subchannel correlations. A major contribution of this paper is that in these measured channels, we observe a coincidence of low K factors and high XPD. In such channels, MIMO systems employing polarization diversity incur SNR and diversity deficits, when compared to spatial configurations. On the other hand, our results indicate that polarization diversity can substantially lower the subchannel correlations for compact configurations, even in a LOS scenario. We draw a fair comparison in terms of capacity, between spatial MIMO configurations and systems using polarization diversity. We analyze the performance of 2 × 2 and 4 × 4 MIMO configurations for a range of values of inter-element spacing.Index Terms-Polarization diversity, MIMO, dual-polarized antennas, inter-element spacing, indoor channels.

Antenna selection combined with dual-polarized antennas offers an attractive alternative for realizing higher order multiple-input multiple-output (MIMO) configurations in compact, low-complexity devices. In this paper we analyze the performance of antenna selection for narrowband dualpolarized MIMO channels with linear minimum mean square error (MMSE) receiver processing. We analytically study the impact of cross-polar discrimination (XPD) on the achieved selection gain. We use channel measurement data, collected at 2.4 GHz in a typical office environment, to compare the performance of spatial and dual-polarized MIMO channels with respect to antenna selection. Our bit-error-rate (BER) results indicate that antenna selection with dual-polarized antennas can achieve significant performance gains for compact configurations.

A novel adaptive space-time Rake receiver is proposed for UMTS frequency division duplex mode based on the adaptive minimum bit error rate (AMBER) criterion. The design of this space-time Rake receiver builds on previous results reported by Chen et al. for temporal or spatial processing, respectively. The original AMBER approach assumes short spreading sequences, i.e. they need to be invariant from symbol to symbol. Here we extend the approach to joint space-time processing and tailor the receiver specifically for a UMTS FDD system with long scrambling sequences in the uplink. The base station receiver uses a twc-or fourelement antenna array.