Spectrum agile radio
A spectrum agile radio having one or more variable digital filters is described. To quickly, yet accurately, retune the digital filter(s) a windowing function is applied to an ideal filter characteristic for each of one or more desired frequency bands to generate filter coefficients. Transitioning between coefficients of a previous filter and a current filter is handled to avoid problems associated with discontinuities in the signal processing.
1. A method for modifying coefficients of a digital filter in a communication node for filtering a signal present in a frequency sub-band the method comprising:
filtering the signal over a frequency band which includes the frequency sub-band;
determining an ideal filter function for the frequency band;
performing an inverse Fourier transform on the ideal filter function to generate initial coefficients;
multiplying the initial coefficients by a windowing function provided for the frequency sub-band to generate final filter coefficients; and
substituting said final filter coefficients for previous coefficients of said digital filter.
2. The method of claim 1 , wherein said windowing function operates to:
select a subset of said initial filter coefficients; and
scale an amplitude of said selected subset of said initial filter coefficients, to generate said final filter coefficients.
3. The method of claim 1 , wherein the steps of determining, performing, multiplying and substituting are repeated when another frequency band, different than said frequency band, is to be processed by said digital filter.
4. The method of claim 1 , wherein said coefficients of said digital filter are modified when another signal is provided in another sub-band of the frequency band, whereby the sub-band start frequency and bandwidth of the another sub-band are different from the sub-band start frequency and bandwidth of the sub-band.
5. The method of claim 4 , wherein the sub-band start frequency and bandwidth of the sub-band is adapted to change for a minimum time slot used for scheduling by the communication network device.
6. The method of claim 1 , wherein the step of substituting further comprises: gradually replacing said previous coefficients with said final filter coefficients.
7. The method of claim 6 , further comprising: replacing one of said previous coefficients with one of said final filter coefficients per signal sample being processed.
8. The method of claim 7 , wherein said signal sample being processed is one of a plurality of zeroed signal samples which are processed through said digital filter during a transition period between using said previous coefficients and said final filter coefficients in said digital filter.
9. The method of claim 1 , wherein said digital filter is disposed in a transmit path in said communication node.
10. The method of claim 1 , wherein said digital filter is disposed in a receive path in said communication node.
11. A communication node comprising:
a plurality of receive/transmit path elements configured to process and filter radio signals, present in a frequency band; and
a processor configured to:
provide a digital filter with a variable set of filter coefficients for filtering a signal present in a sub-band of the frequency band;
modify said variable set of filter coefficients by determining an ideal filter function for the band associated with said radio signal, performing an inverse Fourier Transform on the ideal filter function to generate initial coefficients, multiplying the initial coefficients by a windowing function provided for the frequency sub-band to generate final filter coefficients, and substituting said final filter coefficients for previous coefficients of said at least one digital filter.
12. The communication node of claim 11 , wherein the processor uses said windowing function to select a subset of said initial filter coefficients, and scales an amplitude of said selected subset of said initial filter coefficients, to generate said final filter coefficients.
13. The communication node of claim 11 , wherein the processor adaptively changes the filter coefficients for another signal in another frequency sub-band, different than said sub-band.
14. The communication node of claim 13 , wherein said coefficients of said digital filter are modified, when the sub-band start frequency and bandwidth of the another sub-band change when another signal is provided in another sub-band of the frequency band.
15. The method of claim 14 , wherein the sub-band start frequency and bandwidth of the sub-band is adapted to change for minimum a time slot used by the communication network device.
16. The communication node of claim 11 , wherein the processor is further configured to perform the substituting by gradually replacing said previous coefficients with said final filter coefficients.
17. The communication node of claim 16 , wherein said processor is further configured to replace one of said previous coefficients with one of said final filter coefficients per signal sample being processed.
18. The communication node of claim 17 , wherein said signal sample being processed is one of a plurality of zeroed signal samples which are processed through said digital filter during a transition period between using said previous coefficients and said final filter coefficients in said digital filter.
19. The communication node of claim 11 , wherein said digital filter is disposed in a transmit path in said communication node.
20. The communication node of claim 11 , wherein said digital filter is disposed in a receive path in said communication node.
21. The method of claim 1 , wherein the ideal filter function is determined based on the start frequency and bandwidth of the sub-band.
22. The communication node of claim 11 , wherein the ideal filter function is determined based on the start frequency and bandwidth of the sub-band.