Kaywan H. Afkhamie, Timothy Davidson, Zhi-Quan (Tom) Luo and K. Max Wong.
Constrained adaptive estimation using interior point optimization.
In Proceedings of the 1998 Symposium on Image, Speech,
Signal Processing and Robotics,
volume II, pages 21-26, Hong Kong, Sep. 1998.
Linearly constrained adaptive filtering problems occur in applications such as adaptive digital beamforming. The main drawbacks with existing algorithms are slow convergence in the case of LCMV (Frost, 1972) and numerical stability (RLS based algorithms). In this paper we present a novel approach to the problem using techniques from interior point optimization. The proposed algorithm uses a logarithmic barrier function to compute the new parameter vector. Its convergence speed matches that of the RLS based methods but without the associated problems of numerical instability.
Back to my publications-by-topic page.