What is claimed is:
1. A digital filter, comprising:
an adaptive filter having adaptively variable coefficients;
a time-invariant linear filter generating an error signal for feedback from an error signal generated using an output of the adaptive filter; and
a learning circuit adaptively learning the coefficients of the adaptive filter using an error signal fed back from the time-invariant linear filter.
2. A digital filter, comprising:
an adaptive filter having adaptively variable coefficients;
a first time-invariant linear filter to which an output signal of the adaptive filter is inputted;
a second time-invariant linear filter generating an error signal for feedback from an error signal of an output of the first linear filter; and
a learning circuit adaptively learning the coefficients of the adaptive filter using an error signal fed back from the second linear filter.
3. The digital filter according to claim 2, wherein
said second linear filter has filter coefficients obtained by inverting an order of filter coefficients of the first linear filter.
4. The digital filter according to claim 2, further comprising
a third linear filter generating a signal having a characteristic different from a characteristic of the first linear filter, from the output signal of said adaptive filter.
5. The digital filter according to claim 4, further comprising
a decision circuit detecting a phase error using an output signal of said third linear filter and generating a signal for timing recovery.
6. A digital filter, comprising
an adaptive filter having adaptively variable coefficients;
a decision circuit performing a linear operation using an output signal of the adaptive filter and generating an error signal of an obtained signal;
a time-invariant linear filter generating an error signal for feedback from the error signal generated by the decision circuit; and
a learning circuit adaptively learning the coefficients of the adaptive filter using an error signal fed back from the linear filter.
7. A digital filter, comprising
an adaptive filter having adaptively variable coefficients;
a decision circuit estimating a signal series of an input signal from an output signal of the adaptive filter by one of a maximum likelihood decision and a maximum a posteriori probability decision and generating an error signal; and
a learning circuit learning the coefficients of the adaptive filter using an error signal fed back from the decision circuit in such a way that an expected difference between a sum of squares of an output error signal series for a correct signal series and a sum of squares of an output error signal series in a case of a decision error increases.
8. The digital filter according to claim 7, further comprising
a linear filter having an impulse response corresponding to a self correlation function of an output error signal of said adaptive filter generated when the said decision circuit makes a decision error of a specific pattern and generating an error signal for feedback from the error signal generated by said decision circuit, wherein
said learning circuit learns the coefficients of said adaptive filter using the error signal generated by the linear filter.
9. The digital filter according to claim 8, wherein
the specific pattern is an error pattern in which an error is most easily committed and is determined by an input signal and an ideal output signal of said adaptive filter.
10. The digital filter according to claim 8, wherein
the specific pattern is a single bit error pattern.
11. The digital filter according to claim 8, wherein
the error signal is fed back to said learning circuit from the linear filter only if a signal series after a decision causes the specific pattern.
12. The digital filter according to claim 7, wherein
said learning circuit learns the coefficients of said adaptive filter in such a way that a weighted average of the expected difference between the sums of squares of an error for a plurality of error patterns increases.
13. A digital filter, comprising:
an adaptive filter having adaptively variable coefficients;
a first linear filter to which an output signal of the adaptive filter is inputted;
a decision circuit estimating a signal series of an input signal from an output signal of the first linear filter by one of a maximum likelihood decision and a maximum a posteriori probability decision and generating an error signal;
a second linear filter having an impulse response corresponding to a self correlation function of an output error signal of the adaptive filter generated when the decision circuit makes a decision error of a specific pattern and to which the error signal generated by the decision circuit is inputted;
a third linear filter generating an error signal for feedback from an output signal of the second linear filter; and
a learning circuit learning the coefficients of the adaptive filter using an error signal fed back from the third linear filter in such a way that an expected difference between a sum of squares of an output error signal series for a correct signal series and a sum of squares of an output error signal series in a case of a decision error increases.
14. A digital filter, comprising:
an adaptive filter having adaptively variable coefficients;
a first linear filter to which an output signal of the adaptive filter is inputted;
a decision circuit estimating a signal series of an input signal from an output signal of the first linear filter by one of a maximum likelihood decision and a maximum a posteriori probability decision and generating an error signal;
a second linear filter to which the error signal generated by the decision circuit is inputted;
a third linear filter having an impulse response corresponding to a self correlation function of an output error signal of the adaptive filter generated when the decision circuit makes a decision error of a specific pattern and generating an error signal for feedback from an output signal of the second linear filter;
a learning circuit learning the coefficients of the adaptive filter using an error signal fed back from the third linear filter in such a way that an expected difference between a sum of squares of an output error signal series for a correct signal series and a sum of squares of an output error signal series in a case of a decision error increases.
15. A digital filter, comprising:
an adaptive filter having adaptively variable coefficients;
a first linear filter to which an output signal of the adaptive filter is inputted;
a decision circuit estimating a signal series of an input signal from an output signal of the first linear filter by one of a maximum likelihood decision and a maximum a posteriori probability decision and generating an error signal;
a second linear filter having a transfer characteristic expressed by a product of a filter characteristic obtained by inverting an order of filter coefficients of the first linear filter and an impulse response corresponding to a self correlation function of an output error signal of the adaptive filter generated when the decision circuit commits a decision error of a specific pattern, and generating an error signal for feedback from the error signal generated by the decision circuit;
a learning circuit learning the coefficients of the adaptive filter using an error signal fed back from the second linear filter in such a way that an expected difference between a sum of squares of an output error signal series for a correct signal series and a sum of squares of an output error signal series in a case of a decision error increases.
16. A digital filter, comprising
an adaptive filter having adaptively variable coefficients;
an interpolation filter interpolating an output signal of the adaptive filter and generating a signal in a prescribed timing;
an inverse interpolation filter having filter coefficients obtained by inverting an order of one of interpolation coefficients corresponding to an output of the interpolation filter and approximate values of the interpolation coefficients, and generating an error signal in timing before interpolation by inversely interpolating an error signal of the output of the interpolation filter; and
a learning circuit adaptively learning the coefficients of the adaptive filter using an error signal fed back from the inverse interpolation filter.
17. The digital filter according to claim 16, wherein
said inverse interpolation filter has filter coefficients obtained by inverting an order of a plurality of coefficients of a large absolute value of the interpolation coefficients.
18. The digital filter according to claim 16, wherein
said inverse interpolation filter has filter coefficients obtained by inverting an order of two coefficients of a large absolute value of the interpolation coefficients.
19. The digital filter according to claim 16, wherein
said inverse interpolation filter has filter coefficients obtained by inverting an order of coefficients obtained by linearly approximating two coefficients of a large absolute value of the interpolation coefficients with respect to interpolation phase.
20. A digital filter, comprising:
an adaptive filter having adaptively variable coefficients;
an interpolation filter interpolating an output signal of the adaptive filter and generating a signal in a prescribed timing;
a first linear filter to which an output signal of the interpolation filter is inputted;
a second linear filter to which an error signal of an output of the first linear filter is inputted;
an inverse interpolation filter having filter coefficients obtained by inverting an order of one of interpolation coefficients corresponding to an output of the interpolation filter and approximate values of the interpolation coefficients, and generating an error signal in timing before interpolation by inversely interpolating an output signal of the second linear filter; and
a learning circuit adaptively learning the coefficients of the adaptive filter using an error signal fed back from the inverse interpolation filter.
21. A digital filter, comprising:
an adaptive filter having adaptively variable coefficients;
an interpolation filter interpolating an output signal of the adaptive filter and generating a signal in a prescribed timing;
a decision circuit performing a linear operation using an output signal of the interpolation filter and generating an error signal of an obtained signal;
a linear filter to which the error signal generated by the decision circuit is inputted;
an inverse interpolation filter having filter coefficients obtained by inverting an order of one of interpolation coefficients corresponding to an output of the interpolation filter and approximate values of the interpolation coefficients, and generating an error signal in timing before interpolation by inversely interpolating an output signal of the linear filter; and
a learning circuit adaptively learning the coefficients of the adaptive filter using an error signal fed back from the inverse interpolation filter.
22. A digital filter, comprising:
an adaptive filter having adaptively variable coefficients;
a decision circuit generating an error signal of an output of the adaptive filter;
a linear filter generating an effective input signal from an input signal of the adaptive filter; and
a learning circuit adaptively learning the coefficients of the adaptive filter by correlating the effective input signal from the linear filter to an error signal fed back from the decision circuit.
23. A digital filter, comprising:
an adaptive filter having adaptively variable coefficients;
a decision circuit performing a linear operation using an output signal of the adaptive filter and generating an error signal of an obtained signal;
a linear filter generating an effective input signal from an input signal of the adaptive filter; and
a learning circuit adaptively learning the coefficients of the adaptive filter by correlating the effective input signal from the linear filter to an error signal fed back from the decision circuit.
24. The digital filter according to claim 23, wherein
said linear filter has one of a characteristic equivalent to a remaining part obtained by eliminating a transfer characteristic of the adaptive filter from a transfer characteristic of a route between the input signal of the adaptive filter and the signal obtained by the linear operation, and a characteristic obtained by approximating the remaining part.
25. A digital filter, comprising:
an adaptive filter having adaptively variable coefficients;
a first linear filter to which an output signal of the adaptive filter is inputted;
a decision circuit performing a linear operation using an output signal of the adaptive filter and generating an error signal of an obtained signal;
a second linear filter generating an effective input signal from an input signal of the adaptive filter; and
a learning circuit adaptively learning the coefficients of the adaptive filter by correlating the effective input signal from the second linear filter to an error signal fed back from the decision circuit.
26. A digital filter, comprising:
an adaptive filter having adaptively variable coefficients;
a decision circuit performing a linear operation using an output signal of the adaptive filter and generating an error signal of an obtained signal;
a first linear filter generating an error signal for feedback from the error signal generated by the decision circuit;
a second linear filter generating an effective input signal from an input signal of the adaptive filter; and
a learning circuit adaptively learning the coefficients of the adaptive filter by correlating the effective input signal from the second linear filter to an error signal fed back from the first linear filter.
27. The digital filter according to claim 26, wherein
a product of a transfer characteristic of the first linear filter and a transfer characteristic of the second linear filter is expressed by one of a remaining characteristic obtained by eliminating a transfer characteristic of the adaptive filter from a transfer characteristic of a route between the input signal of the adaptive filter and the signal obtained by the linear operation, and a characteristic obtained by approximating the remaining part.
28. A digital filter, comprising:
a fixed-coefficient filter having filter coefficients determined by a simulation that provides a digital filter model, which learns coefficients of an adaptive filter having adaptively variable coefficients by feeding back an error signal generated using an output of the adaptive filter through a time-invariant linear filter, with an input signal of a specific statistical property and conducts coefficient learning; and
a learning circuit estimating a signal series of an input signal from an output signal of the fixed-coefficient filter.
29. A digital filter, comprising:
an adaptive filter means having adaptively variable coefficients;
a time-invariant linear filter means for generating an error signal for feedback from an error signal generated using an output of the adaptive filter means; and
a learning circuit means for adaptively learning the coefficients of the adaptive filter means using an error signal fed back from the time-invariant linear filter means.
30. A digital filter, comprising
an adaptive filter means having adaptively variable coefficients;
an interpolation filter means for interpolating an output signal of the adaptive filter means and generating a signal in a prescribed timing;
an inverse interpolation filter means having filter coefficients obtained by inverting an order of one of interpolation coefficients corresponding to an output of the interpolation filter means and approximate values of the interpolation coefficients, and generating an error signal in timing before interpolation by inversely interpolating an error signal of the output of the interpolation filter means; and
a learning circuit means for adaptively learning the coefficients of the adaptive filter means using an error signal fed back from the inverse interpolation filter means.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. A method for controlling a hydraulic system adapted to perform at least one hydraulic work function in a working machine (1), wherein the hydraulic system performs the hydraulic control function in accordance with an operation signal (F) determined by a hydraulic system control unit (24), comprising the steps of:
the control unit (24) receiving an operator control input (a) associated to said work function, characterized in that
the control unit receiving a load input (L) indicative of a load associated to the work function,
the control unit determining the operation signal (F) in response to the operator control input (a) and the load input (L).
2. A method according to claim 1 wherein a relationship between the load input (L) and the operation signal (F) is determined so that an increased load input value (L) results in a reduced operation signal value (F).
3. A method according to claim 2 wherein the increased load input value (L) results in a reduced operation signal value (F) within a first control range of the operator control input (a), the first control range being smaller than the total operating range of the control input (a).
4. A method according to any of the claims 1-3 wherein a relationship between the load input (L) and the operation signal (F) is linear over the total operating range of the operator control input (a).
5. A method according to any of the claims 1-3 wherein a relationship between the load input (L) and the operation signal (F) is non-linear over the total operating range of the operator control input (a).
6. A method according to any of the claims 1-3 wherein a relationship between the load input (L) and the operation signal (F) is linear in at least a first interval of the load input value (L) and non-linear in a second interval of the load input value (L) for a particular value of the operator control input (a).
7. A method according to any of the preceding claims wherein a relationship between the load input (L) and the operation signal (F) in terms of gradient or slope is the same for different values of the operator control input (a).
8. A method according to any of the claims 1-6 wherein a relationship between the load input (L) and the operation signal (F) in terms of gradient or slope varies for different values of the operator control input (a).
9. Method according to any of the preceding claims wherein a relationship between the load input (L) and the operation signal (F) is defined in a control map in the hydraulic system control unit (24).
10. A method according to any of the preceding claims wherein the operator control input (a) corresponds to an angle of an operation lever.
11. A method according to any of the preceding claim wherein a relationship between the load input (L) and the operation signal (F) for a particular operator control input value (a) is dependent on an operating state of the working machine (1).
12. A method according to any of the preceding claims wherein the load input (L) is determined on the basis of a load indicating signal.
13. A method according to any of the preceding claims wherein the control unit (25) determines a load change in the work function during operation on the basis of repeated load inputs (L).
14. A method according to claim 13 wherein the control unit (24) changes the desired speed of the work function in response to the determined load change.
15. A hydraulic system control unit (24) being adapted to perform any of the method steps according to any of the claims 1-14.
16. A hydraulic system (17) comprising a hydraulic system control unit (24) according to claim 15.
17. A working machine (1) comprising a hydraulic system according to claim 16.