In a previous work, the principles of the eigenanalysis method for interference cancellation was studied for uniform regular array (URA) structure, wherein the array elements are equally spaced, The main objective of this Paper is to extend the results of that work to adaptive arrays with different geometry termed ''minimum redundancy array'' (MRA). An efficient method is presented to solve the minimum variance optimization problem and explicit analytical solutions are derived for the optimal weight-vector and output noise variance of the eigencanceler. Performance analysis of the general N-element array was first derived and used to extract the performance of MRA. URA is also considered as a special case. The closed-form equation depicts the minimum variance noise residue for the single interference, case, while an interactive procedure is suggested for the two interferences case. If there are more than two interferences, we propose to use direct numerical calculation.