Robust spatial covariance matrix transformation techniques for downlink beamforming

In this paper, we propose robust uplink to downlink spatial covariance matrix transformation techniques for a frequency selective fading environment of a UMTS-FDD system. Our approach leads to a matrix robust least squares (RLS) problem that can be formulated both in a convex form as a second order cone (SOC) program (Method-I) as well as in a regularized least square form (Method-II). The performance of the proposed methods are compared with that of (K. Hugl et al., 1999) and (T. Aste et al., 1998) using a dynamic system level simulator developed for investigating the capacity improvement of the UMTS-FDD system with smart at the base stations. Downlink capacity is used as measure of system performance by finding the number of users for which a certain quality of service (QoS) requirement fulfilled. System level simulations are carried out for 4-element uniform circular antenna (UCA) arrays. Simulation show that robust methods outperform both (K. Hugl et al., 1999) and (T. Aste et al., 1998) by supporting about 50% and 16% more average number of users than that of (K. Hugl et al., 1999) and (T. Aste et al., 1998), respectively without any significant degradation in performance when compared to (T. Aste et al., 1998) with known uplink spatial signature.