Multidimensional autoregression

Test results for leveled inverse interpolation

Figures 26 and 27 show the same example as in Figures and . What is new here is that the proper PEF is not given but is determined from the data. Figure 26 was made with a three-coefficient filter $(1,a_1,a_2)$ and Figure 27 was made with a five-coefficient filter $(1,a_1,a_2,a_3,a_4)$ . The main difference in the figures is where the data is sparse. The data points in Figures , 26 and 27 are samples from a sinusoid.

subsine3 Figure 26. Interpolating with a three-term filter. The interpolated signal is fairly monofrequency.

subsine5 Figure 27. Interpolating with a five term filter.

Comparing Figures and to Figures 26 and 27 we conclude that by finding and imposing the prediction-error filter while finding the model space, we have interpolated beyond aliasing in data space.

Sometimes PEFs enable us to interpolate beyond aliasing.

Multidimensional autoregression