Online signal filtering based on the algebraic method and its experimental validation

• An on-line algebraic filtering scheme has been presented.
• The methodology uses a time derivative estimator and an algebraically based signal filter.
• The methodology presents high noise attenuation and is computed on-line and in real time.
• The approach has high versatility and easy implementation in real engineering applications.
• There are well correspondences between the experimental and simulation results.

An on-line algebraic filtering scheme, based on the recently introduced algebraic approach to parameter and state estimation, is presented along with successful experimental results. The proposed filtering algorithm is based on the connections between a time derivative estimator and an algebraically based signal filtering option. The main advantages of the proposed approach are: (i) there are no appreciable delays in the filtered signal; (ii) the method does not require any statistical assessment of the noises corrupting the signal; (iii) high attenuation of the noise effects is achieved; (iv) the on-line computations are carried out in real time; and (v) high versatility and ease of implementation. Several experiments related to real depth measurements were conducted to show the effectiveness of the proposed algorithm. Comparisons are performed with different filtering alternatives.