Statistical Approach to Basis Function Truncation in Digital Interpolation Filters

In this paper an alternative analysis in the time domain is described and the results of the interpolation process are presented by means of functions that are based on the rule of conditional mathematical expectation and the covariance function. A comparison between the interpolation error caused by low order filters and the classic sinc(t) truncated function is also presented. When fewer samples are used, low-order filters have less error. If the number of samples increases, the sinc(t) type functions are a better alternative. Generally speaking there is an optimal filter for each input signal which depends on the filter length and covariance function of the signal. A novel scheme of work for adaptive interpolation filters is also presented.

Authors:

• F. Castillo
• J. Arellano
• S. Sánchez

Keywords:

• Interpolation
• basis function
• over-sampling.

References:

[1] Farokh Marvasti (ed.), Nonuniform Sampling, Theory and Practice
(Kluwer Academic/Plenum Publishing, 2001).
[2] S. Mitra, Digital Signal Processing: A Computer-Based Approach,
3rd edition. (McGraw-Hill International Press, 2005).
[3] Steven W. Smith, The Scientist and Engineer-s Guide to Digital
Signal Processing, 2nd edition (California Technical Publishing,
1999).
[4] Analog Devices Engineering Staff, Mixed-Signal and DSP Design
Techniques, ed. Walt Kester (USA, 2000).
[5] Thomas Strohmer, Implementations of Shannon's Sampling Theorem:
A Time-Frequency Approach. Sampling Theory in Signal and Image
Processing (Sampling Publishing, Jan. 2005).
[6] R. L. Stratonivich, Topics in the Theory of Random Noise (Gordon
and Breach, 1963).
[7] V. A. Kazakov, Sampling-Reconstruction Procedure of Gaussian
Fields. Abstracts of the International Conference ''Sampling Theory
and Applications'' (SAMPTA 2003), (Strobl, Salzburg, Austria, May
2003).
[8] Vladimir Kazakov and Sviatoslav Afrikanov, Sampling -
Reconstruction Procedure of Gaussian Fields, Computaci├│n y
Sistemas, CIC-IPN, 9/3 (2006).
[9] Daniel Saldaña Rodríguez, Estudio del procedimiento de muestreoreconstrucción
de los procesos Gaussianos con una cantidad de
muestras limitada (Master-s thesis, ESIME-IPN, México DF, 2003).
[10] I. Mednieks, Methods For Spectral Analysis Of Nonuniformly
Sampled Signals. (International Conference on Sampling Theory and
Application, SAMPTA 2005).
[11] Jes├║s Bernal Berm├║dez and Jes├║s Bobadilla, Reconocimiento de voz
y fonética ac├║stica (Alfaomega, Espa├▒a, 2003).
[12] Saeed Mian Qaisar and Laurent Fesquet. Adaptive Rate Sampling and
Filtering for Low Power Embedded Systems (International
Conference on Sampling Theory and Application, SAMPTA 2007).
[13] National Semiconductor, ADC16071/ADC16471, 16-Bit Delta-Sigma
192 ks/s Analog-to-Digital Converters (Datasheet, 1995).
[14] A. V. Oppenheim and R. W. Schafer, with J. R. Buck, Discrete-Time
Signal Processing, 2nd edition (Prentice-Hall, Inc., Upper Saddle
River, NJ, 1999).
[15] Cormac Herley and Ping Wah Wong, Minimum Rate Sampling and
Reconstruction of Signals with Arbitrary Frequency Support. (IEEE
Transactions on Information Theory, July 1999).