1: \begin{abstract}

2: We introduce a novel family of adaptive filtering algorithms based on

3: a relative logarithmic cost. The new family intrinsically combines the

4: higher and lower order measures of the error into a single continuous

5: update based on the error amount. We introduce important members of

6: this family of algorithms such as the least mean logarithmic square

7: (LMLS) and least logarithmic absolute difference (LLAD) algorithms

8: that improve the convergence performance of the conventional

9: algorithms. However, our approach and analysis are generic such that

10: they cover other well-known cost functions as described in the

11: paper. The LMLS algorithm achieves comparable convergence performance

12: with the least mean fourth (LMF) algorithm and extends the stability

13: bound on the step size. The LLAD and least mean square (LMS)

14: algorithms demonstrate similar convergence performance in impulse-free

15: noise environments while the LLAD algorithm is robust against

16: impulsive interferences and outperforms the sign algorithm (SA). We

17: analyze the transient, steady state and tracking performance of the

18: introduced algorithms and demonstrate the match of the theoretical

19: analyzes and simulation results. We show the extended stability bound

20: of the LMLS algorithm and analyze the robustness of the LLAD algorithm

21: against impulsive interferences. Finally, we demonstrate the

22: performance of our algorithms in different scenarios through numerical

23: examples.

24: \end{abstract}