Linear Programming Applied to the Optimization of Blind Adaptive Antenna Arrays

Arrays of adaptive antennas are widely used in wireless communication systems, where the array coefficients are adjusted in real time in order to reduce the rate of data errors due to interferences. Adjustment of the coefficient can be performed using supervised or blind methods. In the case of the latter, it is also possible to improve the transmission capacity of the system. Nevertheless, the blind algorithms used to optimize the array coefficients employ nonconvex functions, which reduce their efficiency in the presence of local minima. Different to the solutions that have been suggested in the literature, this article presents a proposal for convex optimization, based on linear programming, applied to optimization of the coefficients associated with arrays of adaptive antennas. The work also presents a comparative performance evaluation of interior-point, active-set, and simplex methods applied in the optimization process. Simulation results for digital communication systems are presented using bit error rate performance curves.