An adaptive algorithm for structurally passive, VLSI implemented and recursive filters

A least-mean-square-based adaptive algorithm for structurally passive, VLSI implemented, and recursive filters is presented. The lattice coefficients of the structurally passive filter are expressed in terms of trigonometric functions. The coefficients are then updated in such a way that the cost function is minimized with respect to the arguments of the trigonometric functions. This approach makes stability highly tractable during the adaptation process. Moreover, the resulting adaptive filter is of low-sensitive and modular structure. Simulation results are included to demonstrate the algorithm's capability in passive system identification.<>