2009 American Control Conference 2009
DOI: 10.1109/acc.2009.5160354
|View full text |Cite
|
Sign up to set email alerts
|

Are anti-aliasing filters really necessary for sampled-data control?

Abstract: In the paper the role of anti-aliasing filters is revised based on control quality assessment of various setups of analog filters used prior to sampling, or their lack. Numerical results show that contrary to common belief possible benefits gained from anti-aliasing filters are very restricted.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1

Citation Types

1
4
0

Year Published

2009
2009
2017
2017

Publication Types

Select...
5
1

Relationship

2
4

Authors

Journals

citations
Cited by 9 publications
(5 citation statements)
references
References 5 publications
1
4
0
Order By: Relevance
“…The best results in the latter case are obtained when the continuous-time output is passed through a continuous-time Kalman filter before being sampled. Similar results were observed in PID control systems [6], [7], [8] In the present paper we study these issues in the context of optimal discrete-time PID control and compare the results with LQG minimum-variance control.…”
Section: Introductionsupporting
confidence: 78%
See 1 more Smart Citation
“…The best results in the latter case are obtained when the continuous-time output is passed through a continuous-time Kalman filter before being sampled. Similar results were observed in PID control systems [6], [7], [8] In the present paper we study these issues in the context of optimal discrete-time PID control and compare the results with LQG minimum-variance control.…”
Section: Introductionsupporting
confidence: 78%
“…In [4], [5] we studied the impact of antialiasing filters for pure signal processing while in [7] in the discrete-time LQG control context, and we have found that there is no reason for using them in the noiseless case, and practically no use in the case of noisy measurements. The best results in the latter case are obtained when the continuous-time output is passed through a continuous-time Kalman filter before being sampled.…”
Section: Introductionmentioning
confidence: 98%
“…Anti-aliasing filters usually take the form of Butterworth filters whose cutoff frequency equals to the so called Nyquist frequency ω N = π/h depending solely on sampling period h. As an alternative, so called integrating or averaging samplers are considered in [4,5,7,8]. In [4,5] we studied the impact of antialiasing filters for pure signal processing while in [6] in the discrete-time LQG control context, and we have found that there is no reason for using them in the noiseless case, and practically no use in the case of noisy measurements. The best results in the latter case are obtained when the continuous-time output is passed through a continuous-time Kalman filter before being sampled.…”
Section: Introductionmentioning
confidence: 99%
“…As is well known, any signal content above the Nyquist sampling rate become aliased. The standard approach of adding an analog anti-aliasing filter does not always improve performance 35,36 . The advantage of rolling off noise and resonant dynamics must be weighed against the added phase delay to the closed loop system and the loss in performance from this lag.…”
Section: Digital Control Design With An Fpgamentioning
confidence: 99%