Adaptive arrays can significantly increase cell capacity, improve signal quality, and reduce transmitter power requirements. In this article we establish a relationship between the information theoretic capacity of a mobile radio system and the beam pattern of a multisensor array. We investigate the capacity improvement potentially achieved via an optimized design strategy for an unequally spaced array; that is, the positions and the weighting coefficients of the array elements are selected to improve the average system capacity subject to various constraints like minimizing the maximum sidelobe level or keeping the beamwidth of the main lobe to a minimum. Next, we investigate the effect of fading correlation on the performance of an unequally spaced adaptive array. Results are presented for optimum combining with flat fading as well as for frequency-selective fading using a two-path delay spread model. Computer simulations show that it is possible to achieve a gain of 1.5 dB for moderate to high signal-to-noise ratios when compared to the equally spaced array. Finally, it is shown that a base station with wide antenna element spacing has improved bit error rate performance over one with narrow element spacing under cochannel interference and multipath fading. In particular, we improve the performance for both uplink and downlink transmissions in a slow fading channel with cochannel interference. Promising results are presented for a 4 channel carrier time division duplexing system.