Sliding window order-recursive least-squares algorithms

Order-recursive least-squares (ORLS) algorithms employing a sliding window (SW) are presented. The authors demonstrate that standard architectures that are well known for growing memory ORLS estimation, e.g., triangular array, lattice, and multichannel lattice, also apply to sliding window ORLS estimation. A specific SW-ORLS algorithm is the combination of two independent attributes: its global architecture and its local cell implementation. Various forms of local cell implementation based on efficient time-recursions of time-varying coefficients are discussed. In particular, the authors show that time and order updates of any order-recursive sliding window least-squares algorithm can be realized solely in terms of 3×3 hyperbolic Householder transformations (HHT). Finally, the authors present two HHT-based algorithms: the HHT triangular array algorithm and the HHT lattice algorithm