Steering an adaptive antenna array by the simplified Kalman filter

Adaptive antenna arrays controlled by the Kalman filter excel in high rate of convergence, low misadjustment and independence on a priori information about interference statistics. High computational requirements and potential numerical non-stability are drawbacks of the Kalman´s steering. The author describes a way of reducing computational requirements and improving numerical stability of the steering algorithm based on a Kalman filter. Rate of convergence and misadjustment of the resultant algorithm (simplified Kalman filter, (SKF)) are better in comparison with the optimal LMS and worse than the pure Kalman filter. Computational requirements of the SKF are approximately three times higher in comparison with the LMS. SKF does not require any a priori information about interference statistics and is totally numerically stable.