Performance of a high-precision digital tuning algorithm

Abstract

An iterative digital tuning algorithm has been reviewed and simulated extensively, and its functionality has been investigated experimentally. This tuning algorithm is applicable to digitally programmable continuous-time filters. The algorithm estimates filter characterization parameters based on a system identification method and calculates control parameters for tuning adjustments. Extensive simulations show that this algorithm converges to a solution within 10 iterations in most cases, even in the presence of various measurement errors, parameter variations, and overordering effects. This algorithm is very insensitive to initial parameter variations, as good results are obtained even with 30% parameter error. This algorithm also attains good speed and accuracy in the presence of 5% measurement errors and high overordering effects. Experimental results with a second-order monolithic programmable OTA-C (operational transconductance amplifier and capacitor) integrated circuit filter verify the robustness of the algorithm in tuning continuous-time filters.<>