A sampling technique for non-bandlimited signals based on finite update rate
Liping Jiang · Journal of Circuits and Systems · 2008
Aiming at the drawback that Shannon theory can only process bandlimited signals, this paper deduces a kind of sampling theory of non-bandlimited signals based on finite update rate. The signal, which has finite update rate, is filtered through a sample kernel and then is sampled uniformly at an appropriate rate which is not less than the update rate. Then the original signal could be perfectly reconstructed by virtue of annihilating filter and Vandermonde equation. The simulation results indicate that this algorithm has high precision and good anti-noise performance. Its mean square error is only 10-5 level when the signal noise rate is 20dB, which can satisfy the engineering requirements.