Adaptive switched capacitor filters based on the LMS algorithm

An approach to implementing an adaptive FIR (finite-impulse response) filter based on the LMS (least-mean-square) algorithm using SC (switched-capacitor) circuits is described. A concept for such a realization is presented, consisting of the following basic elements: delay element, summer, integrator, and multiplier. The influence of nonideal effects of SC networks on the behavior of the filter is investigated in an error analysis. It is shown that nonidealities in the FIR filter part can be eliminated by an additional constant tap element, whereas the main error source in the adaptation part is the integrator offset errors, which can be eliminated by offset-free SC integrators.<<ETX>>