Sampled data digital filtering system

1. A digital filter for filtering sample data, comprising:

a delay network for delaying input sample data to provide a plurality of delayed sample data outputs;

a filter network representable by a decomposed coefficient weighting matrix for processing said delayed sample data outputs; and

a processor for producing a filtered output by computing a weighted product summation of said delayed sample data outputs and said coefficient weighting matrix, said processor being responsive to a sample spatial position index signal in producing said filtered output.

2. A digital filter according to claim 1 , wherein

said decomposed coefficient weighting matrix comprises a structurally factored matrix.

3. A digital filter according to claim 2 , wherein

said structurally factored matrix employs a factor derived based on a property including at least one of, (a) coefficient matrix row symmetry, and (b) coefficient matrix column symmetry.

4. A digital filter according to claim 1 , wherein

said decomposed coefficient weighting matrix is derived by at least one of (a) factoring a first coefficient weighting matrix with a common row factor, and (b) factoring a first coefficient weighting matrix based on at least one of, (i) coefficient matrix row symmetry, and (ii) coefficient matrix column symmetry.

5. A digital filter according to claim 1 , wherein

said decomposed coefficient weighting matrix is derived by factoring a first coefficient weighting matrix using a sparse matrix.

6. A digital filter according to claim 1 , wherein

said decomposed coefficient weighting matrix represents a multiple input, multiple output, filter network.

7. A digital filter according to claim 1 , including

an interpolation network for interpolating sample data to provide said input sample data.

8. A digital filter according to claim 1 , wherein

said processor includes a factor combiner for deriving a weighted sum of factors representing a linear transform process.

9. A digital filter according to claim 1 , wherein

said decomposed coefficient weighting matrix exhibits the form

[

0

0

3

0

-

1

4

-

2

-

1

1

-

1

-

1

1

]

3

.

10. A digital filter according to claim 1 , wherein

said digital filter provides the function

H

⁡

(

z

)

=

[

1

μ

μ

2

]

·

[

0

0

3

0

-

1

4

-

2

-

1

1

-

1

-

1

1

]

3

·

[

1

z

-

1

z

-

2

z

-

3

]

 where u is a sample spatial position representative signal and z represents an input sample.

11. A digital filter according to claim 1 , wherein

said decomposed coefficient weighting matrix exhibits the form

[

6

58

58

6

23

59

-

59

-

23

31

-

31

-

31

31

16

-

48

48

-

16

]

128

.

12. A digital filter according to claim 1 , wherein

said digital filter provides the following function, where u is a sample spatial position representative signal and z presents an input sample

H

⁡

(

z

)

=

[

1

μ

μ

2

μ

3

]

⁡

[

1

2

0

3

64

0

0

1

0

23

128

0

0

31

128

0

0

0

0

1

8

]

·

[

0

1

1

0

0

1

-

1

0

1

-

1

-

1

1

1

-

3

3

-

1

]

·

[

1

z

-

1

z

-

2

z

-

3

]

13. A method for filtering sample data, comprising the steps of:

delaying input sample data to provide a plurality of delayed sample data outputs;

processing said delayed sample data outputs using a filter network represented by a structurally factored coefficient weighting matrix, said structurally factored matrix comprising a coefficient weighting matrix employing factors derived based on a property including at least one of, (a) coefficient matrix row symmetry, and (b) coefficient matrix column symmetry; and

producing a filtered output by computing a weighted product summation of said delayed sample data outputs and said coefficient weighting matrix.

14. A method for filtering sample data, comprising the steps of:

delaying input sample data to provide a plurality of delayed sample data outputs;

processing said delayed sample data outputs using a filter network represented by a structurally factored coefficient weighting matrix, said structurally factored matrix being derived by at least one of (a) factoring a first coefficient weighting matrix with a common row factor, and (b) factoring a first coefficient weighting matrix based on a property including at least one of, (i) coefficient matrix row symmetry, and (ii) coefficient matrix column symmetry; and

producing a filtered output by computing a weighted product summation of said delayed sample data outputs and said coefficient weighting matrix.

15. A method for filtering sample data, comprising the steps of:

delaying input sample data to provide a plurality of delayed sample data outputs;

processing said delayed sample data outputs using a filter network represented by a structurally factored coefficient weighting matrix, said structurally factored matrix comprising a decomposed coefficient weighting matrix derived by factoring a first coefficient weighting matrix using a sparse matrix; and

producing a filtered output by computing a weighted product summation of said delayed sample data outputs and said coefficient weighting matrix.

16. A method for filtering sample data, comprising the steps of:

delaying input sample data to provide a plurality of delayed sample data outputs;

processing said delayed sample data outputs using a filter network using a coefficient weighting matrix comprising,

[

0

0

3

0

-

1

4

-

2

-

1

1

-

1

-

1

1

]

3

;

and

producing a filtered output by computing a weighted product summation of said delayed sample data outputs and said coefficient weighting matrix.

17. A method for filtering sample data, comprising the steps of:

delaying input sample data to provide a plurality of delayed sample data outputs;

processing said delayed sample data outputs using a filter network using a coefficient weighting matrix comprising,

[

6

58

58

6

23

59

-

59

-

23

31

-

31

-

31

31

16

-

48

48

-

16

]

128

;

and

producing a filtered output by computing a weighted product summation of said delayed sample data outputs and said coefficient weighting matrix.