1: \begin{abstract}

2: This paper presents reduced-rank linearly constrained minimum

3: variance (LCMV) beamforming algorithms based on joint iterative

4: optimization of filters. The proposed reduced-rank scheme is based

5: on a constrained joint iterative optimization of filters according

6: to the minimum variance criterion. The proposed optimization

7: procedure adjusts the parameters of a projection matrix and an

8: adaptive reduced-rank filter that operates at the output of the

9: bank of filters. We describe LCMV expressions for the design of

10: the projection matrix and the reduced-rank filter. We then

11: describe stochastic gradient and develop recursive least-squares

12: adaptive algorithms for their efficient implementation along with

13: automatic rank selection techniques. An analysis of the stability

14: and the convergence properties of the proposed algorithms is

15: presented and semi-analytical expressions are derived for

16: predicting their mean squared error (MSE) performance. Simulations

17: for a beamforming application show that the proposed scheme and

18: algorithms outperform in convergence and tracking the existing

19: full-rank and reduced-rank algorithms while requiring

20: comparable complexity.\\

21: 

22: \begin{keywords}

23: Adaptive filters, beamforming, constrained optimization, iterative

24: methods.

25: \end{keywords}

26: \end{abstract}