This paper experimentally studies and models the angular-delay power spectrum density at the mobile station based on the site-specific measurement in a macrocell in urban area of Tokyo. The authors first show the azimuth power spectral density at the mobile station. It is decomposed into the "classes" which represent specific contributions within limited azimuth range, as well as the residual. The site-specific propagation mechanism of the classes are next discussed. Finally, the angular-delay PSD models of both classes and residual are proposed and verified. The analysis and modeling in this paper are antenna independent with the full polarimetric information. Consequently, the results are useful to evaluate the performance of arbitrary array antennas with mixed polarization. Due to the rare number of antenna-independent and full-polarimetric measurements, the significant contribution of the angular-delay PSD channel model can be expected.

The combination of Multiple-Input Multiple-Output (MIMO) and Orthogonal Frequency Division Multiplexing (OFDM) technologies gives wireless communications systems the advantages of lower bit error rate (BER) and higher data rate in frequency-selective fading environments. However, the main drawbacks of MIMO systems are their high complexity and high cost. Therefore, antenna selection in MIMO systems has been shown to be an effective way to overcome the drawbacks. In this paper, we propose two receive antenna selection methods for a MIMO-OFDM system with radio frequency (RF) switches and polarization antenna elements at the receiver side, taking into consideration low computational complexity. The first method selects a set of polarization antenna elements which gives lower correlation between received signals and larger received signal power, thus achieves a lower BER with low computational complexity. The second method first selects a set of polarization antenna elements based on the criterion of the first method and another set of polarization antenna elements based on the criterion of minimizing the correlation between the received signals; it then calculates the signal-to-interference-plus-noise power ratio (SINR) of the two sets and selects a set with larger SINR. As a result, the second method achieves a better BER than the first one but it also requires higher computational complexity than the first one. We use the measured channel data to evaluate the performance of the two methods and show that they work effectively for the realistic channel.