The effect of mutual coupling on the performance of adaptive antennas has been a topic of considerable interest for the last three decades. The general conclusion of the work reported in the open literature is that mutual coupling degrades the performance of adaptive antennas. We have carried out an in-depth study of the effects of mutual coupling on the performance of adaptive antennas. Our studies show that this conclusion is not entirely correct. Yes, one does need the in-situ array manifold to obtain the fi xed response in the desired signal direction. Otherwise, adaptive weights can also suppress the desired signal. Note that for adaptive antennas based on minimizing the mean squared error between the array output and a locally generated reference signal, this is not an issue. However, mutual coupling between antenna elements hardly affects the nulling performance of adaptive antennas. In fact, in a given size aperture, as the number of antenna elements is increased, one obtains better nulling performance, irrespective of the increased mutual coupling between antenna elements. Also, as expected, for strong wideband interfering signals, one should carry out space-time adaptive processing (STAP).A daptive-antenna array systems constitute a multidisci pline technology area that has spanned a number of dec ades in the engineering and scientifi c community, due to advances in electromagnetics and signal processing [1][2][3][4][5][6][7][8]. Adaptive arrays consist of antenna elements that sense the signal environment, and a real-time signal processor that automatically optimizes a user-defi ned metric, such as signal-to-noise ratio (SNR), by combining the weighted output of each element (or channel). Adaptive antennas thus can provide real-time adaptive nulling and beamforming. They have appli cations in radar, navigation, RF communication systems, seismic analysis, and sonar [8].The origins of adaptive nulling can be found in the 1960s, when a sidelobe canceller was developed to aid in radar applications [9], and the least-mean-square (LMS) error algorithm was established based on the steepest-descent method [10]. Both of these contributions enabled automatic interference suppression without knowledge of the interferers. Therefore, a weak desired signal can be extracted from a environment with strong interferers [11]. Experimental verifi cation of adaptive nulling was given in [12,13]. Over the decades, the research expanded to provide improvement in convergence, tracking, robustness, power consumption, and versatility of the signal-processing algorithms [2,6,14,15]. In addition, research in adaptive array applications was extended in the areas of radar [16,17] and varying communications systems, such as spread spectrum, TDMA, CDMA, TOA, and indoor radio [18][19][20][21][22][23][24][25]. A third major area of research for adaptive arrays is the effect of the antenna array design itself on its performance in the adaptive mode [26][27][28][29][30][31][32][33][34][35][36][37]. The element pattern [26,27,31], distrib...