Error criteria analysis and robust data fusion

Except the well-known least mean square error criterion which results in the arithmetic mean operation and the mean absolute value criterion which results in the median operation, other nonlinear operations have not been extensively studied in terms of error criteria in the existing literature. In this paper, the error criteria of the harmonic mean, geometric mean, as well as some other nonlinear operations have been derived. Both simulation and experimental data presented in this paper indicate that these nonlinear operations are more robust than that of the linear arithmetic mean. Therefore, the error criteria analysis method can be used as a general guide line for designing the most appropriate operations for the robust fusion problems.<>