On predicting a band-limited signal based on past sample values

  title={On predicting a band-limited signal based on past sample values},
  author={P. P. Vaidyanathan},
  journal={Proceedings of the IEEE},
  • P. Vaidyanathan
  • Published 1 August 1987
  • Computer Science
  • Proceedings of the IEEE
The prediction of samples of a band-limited real signal xa(t) from a finite number of past samples is considered. It is shown how a signal-independent linear predictor of finite order can be constructed based on Chebyshev polynomials, such that the prediction error tends to zero for sampling rate exceeding the Nyquist rate. 

Figures from this paper

On predictors for band-limited and high-frequency time series
Predictors for Discrete Time Processes With Energy Decay on Higher Frequencies
  • N. Dokuchaev
  • Physics
    IEEE Transactions on Signal Processing
  • 2012
A family of one-step-ahead predictors is suggested for processes of which the energy decays at higher frequencies for which the prediction error can be made arbitrarily small.
On Causal Estimation From Bandlimited Stationary Sequences
  • V. Pohl
  • Mathematics
    IEEE Transactions on Information Theory
  • 2011
The paper derives the transfer function of such an estimation filter, discusses its stability behavior, and applies the result to the causal reconstruction of deterministic signals from its samples.
The predictability of band-limited, high-frequency and mixed processes in the presence of ideal low-pass filters
Pathwise predictability of continuous-time processes is studied in a deterministic setting. We discuss uniform prediction in some weak sense with respect to certain classes of inputs. More precisely,
Limited memory predictors based on polynomial approximation of periodic exponents
It is shown that processes with exponentially decaying Fourier transforms are predictable with these predictors in some weak sense, meaning that convolution integrals over the future times can be approximated by causal convolutions over past times.
Predictors for high frequency processes based on rational polynomials approximation of periodic exponentials
—The paper presents linear integral predictors for continuous time high-frequency processes with a a finite spectrum gap. The predictors are based on approximation of complex valued periodic
Some results on the linear prediction of bandlimited fading processes
  • Y. Xie, C. Georghiades
  • Computer Science
    2004 IEEE 59th Vehicular Technology Conference. VTC 2004-Spring (IEEE Cat. No.04CH37514)
  • 2004
The design and performance analysis of linear prediction of a bandlimited complex Gaussian fading process and the design of robust prediction filters whose coefficients are independent of the second order statistics of the fading process are studied.
3 0 A pr 2 00 8 Predictability of band-limited , high-frequent , and mixed processes in the presence of ideal low-pass filters
We study pathwise predictability of continuous time processes in deterministic setting, when the convolution integrals over future time can be approximated by integrals over past time. Uniform
On recoverability of discrete time signals from sparse observations.
It is shown that there exist wide classes of signals that are everywhere dense in the space of square-summable signals and such that signals from these classes feature robust linear recoverability of their finite traces under very mild restrictions on the location of the observed data.
3 A ug 2 00 7 Predictability of band-limited and other processes
Pathwise predictability of continuous time processes is studied. We found that all band-limited processes are predictable in certain weak sense. More precisely, an integral over future time can be


On the prediction of a band-limited signal from past samples
An alternate proof is given of a recent result due to A. Papoulis on predicting the current value of a wide-sense stationary band-limited random process in terms of its past samples. The possibility
A note on the predictability of band-limited processes
It is shown that a band-limited process can be expressed in terms of its past samples only, provided that the sampling rate is larger than but can be chosen arbitrarily close to the Nyquist rate.
Comments on "A note on the predictability of band-limited processes"
A much simpler proof is presented which covers random and deterministic signals with uniform or nonuniform sampling and has been proved by others using different methods.
Digital filters with equiripple or minimax responses
Techniques for determining the coefficients of digital filters which have equiripple or minimax errors are reviewed and occasionally extended. These techniques include: 1) mapping to provide
A separable Hartley-like transform in two or more dimensions
This letter introduces a discrete, separable, real-to-real transform, called the cas-cas transform. Theorems for the two-dimensional (2-D) case are presented, and the cas-cas transform is compared to
Digital transmission in the subscriber loop
Major developments by the telephone industry have launched a worldwide effort to convert the network into an integrated services digital network (ISDN). The digital transmission issues are explored
Difference Methods and Round-off Error Bounds for the Prediction of Bandlimited Functions from Past Samples
La valeur d'un signal a bande limitee a energie finie peut etre predite a partir d'echantillons d'ensembles discrets
PCM and data transmission codes