Method for automatically controlling the bandwidth of a digital filter and adaptive filter utilizing same

• US 5,371,695 A
• Filed: 10/14/1993
• Issued: 12/06/1994
• Est. Priority Date: 10/14/1993
• Status: Expired due to Fees

• Current Assignee: Visteon Global Technologies Incorporated (Visteon Corporation)
• Original Assignee: Ford Motor Company
• Inventors: Baraszu, Robert C.
• Primary Examiner(s): Mai, Tan V.
• Application Number:
  US08/136,065
• Time in Patent Office: 418 Days
• Field of Search: 364/724.19, 364/724.01, 364/807, 395/3, 395/900
• US Class Current: 708/322
• CPC Class Codes: H03H 21/0012 Digital adaptive filters

Abstract

A method and an adaptive filter are provided wherein the bandwidth of the filter is varied as a function of current and past input signal values based on the amount of noise judged to be present. A first order digital low pass filter is described by the following equation:

y(k)=α(x(k)+(k+1))+(1-2α)y(K-1)

where α is a variable control signal which controls the cutoff frequency of the filter. Fuzzy logic or rules are used to determine α. To determine whether the current input signal is most likely a true signal or noise, three differences are taken between the current and past three input sample signals. Based on a comparison of these differences, α is assigned a value from 0.0 to 0.5 in accordance with the fuzzy rules. If the rules indicate that noise is predominant, the bandwidth of the filter is minimized by setting α=0.0 to suppress the noise. If the fuzzy rules indicate that the current input signal is valid, then the bandwidth of the filter is maximized by setting α=0.5. Values of α between 0.0 and 0.5 reflect an indeterminate judgment by the fuzzy rules.

Claims

1. A method for automatically controlling the bandwidth of a digital filter in response to changes of condition of an input signal to the digital filter, the method comprising the steps of:

determining changes of condition of the input signal;

defining fuzzy rules between the changes of condition of the input signal and a control signal;

modifying the control signal based on the changes of condition and the fuzzy rules; and

utilizing the modified control signal to control the bandwidth of the digital filter in real time so that the digital filter achieves good transient response and noise suppression.

2. The method of claim 1 wherein the input signal is a mass airflow signal.

3. The method of claim 1 wherein the changes of condition include two consecutive changes in the value of the input signal.

4. The method of claim 1 wherein the changes of condition includes three consecutive changes in the value of the input signal.

5. The method of claim 1 wherein the changes in condition include a change in frequency of the input signal.

6. The method of claim 1 wherein the changes in condition include two consecutive changes in the value of the input signal and wherein the step of defining includes the steps of:

defining a first fuzzy rule between a first pair of the three consecutive changes in the value of the input signal and a first intermediate control signal;

defining a second fuzzy rule between a second pair of the three consecutive changes in the value of the input signal and a second intermediate control signal; and

defining a control fuzzy rule between the first and second intermediate control signals and the control signal to the digital filter.

7. The method as claimed in claim 1 wherein the changes in condition include three consecutive changes in the value of the input signal and wherein the step of defining includes the steps of:

defining a first fuzzy rule between a first pair of the three consecutive changes in the value of the input signal and a first intermediate control signal;

defining a second fuzzy rule between a second pair of the three consecutive changes in the value of the input signal and a second intermediate control signal;

defining a third fuzzy rule between a third pair of the three consecutive changes in the value of the input signal and a third intermediate control signal; and

defining a control fuzzy rule between the first, second and third intermediate control signals and the control signal to the digital filter.

8. An adaptive filter having a bandwidth responsive to changes of condition of an input signal, the adaptive filter comprising:

means for determining changes of condition of the input signal;

a digital filter;

means for defining fuzzy rules between the changes of condition of the input signal and a control signal; and

means for modifying the control signal based on the changes of condition and the fuzzy rules wherein the modified control signal controls the bandwidth of the digital filter in real time so that the digital filter achieves good transient response and noise suppression.

9. The adaptive filter of claim 8 wherein the input signal is a mass airflow signal.

10. The adaptive filter of claim 8 wherein the digital filter is a lowpass filter.

11. The adaptive filter of claim 8 or claim 10 wherein the digital filter is a first order filter.

12. The adaptive filter of claim 8 wherein the changes of condition include two consecutive changes in value of the input signal.

13. The adaptive filter of claim 8 wherein the changes of condition include three consecutive changes in the value of the input signal.

14. The adaptive filter of claim 8 wherein the changes in condition include a change in frequency of the input signal.

15. The adaptive filter of claim 8 wherein the changes in condition include two consecutive changes in the value of the input signal and wherein the means for defining include:

means for defining a first fuzzy rule between a first pair of the three consecutive changes in the value of the input signal and a first intermediate control signal;

means for defining a second fuzzy rule between a second pair of the three consecutive changes of the value of the input signal and a second intermediate control signal; and

means for defining a control fuzzy rule between the first and second intermediate control signals and the control signal to the digital filter.

16. The adaptive filter as claimed in claim 8 wherein the changes in condition include three consecutive changes in the value of the input signal and wherein the means for defining includes:

means for defining a first fuzzy rule between a first pair of the three consecutive changes in the value of the input signal and a first intermediate control signal;

defining a second fuzzy rule between a second pair of the three consecutive changes in the value of the input signal and a second intermediate control signal;

means for defining a third fuzzy rule between a third pair of the three consecutive changes in the value of the input signal and a third intermediate control signal; and

defining a control fuzzy rule between the first, second and third intermediate control signals and the control signal to the digital filter.