A systolic array for recursive least squares estimation by inverse updates

We derive a new systolic array for implementing a recursive least squares (RLS) algorithm based on the QR decomposition (QRD) inverse-updates method. Unlike most previous algorithms based on the QRD technique, this algorithm explicitly produces the transversal filter weights. In its original formulation, the QRD inverse-updates algorithm contains a long data feedback path which makes the construction of a systolic architecture very difficult. Here we show how to transform the algorithm so as to allow two systolic implementations: one of which has been published and a hitherto unknown structure.