Granted Patent B2

US 7,158,793 · App. 10/399,108 · Granted Jan 2, 2007

Abstract

Claims (48)

1. A method for beam shaping in a radio communication system having a plurality of subscriber stations and a base station which has an antenna apparatus with a plurality of antenna elements, comprising:

transmitting a downlink signal to a subscriber station from each antenna element;

for each of the antenna elements, weighting the downlink signal transmitted thereby with a weighting coefficient w i , i=1, . . . , M, the weighting coefficients together forming a current weighting vector;

forming a plurality of first weighting vectors in an initialization phase, the first weighting vectors being formed so as to be orthogonal to one another:

cyclically redefining the current weighting vector for each time slot of the downlink signal, the current weighting vector being cyclically redefined based on the first weighting vectors;

producing a first spatial covariance matrix of the downlink signal;

determining eigenvectors of the first spatial covariance matrix; and

selecting the first weighting vectors from among the eigenvectors of the first spatial covariance matrix, wherein

short-term covariance matrices are formed for each tap of the downlink signal; and

the first spatial covariance matrix is obtained by averaging the short-term covariance matrices over all taps.

2. The method as claimed in claim 1 , wherein the first weighting vectors are formed on the basis of measurements of the downlink signal.

3. The method as claimed in claim 1 , wherein the first weighting vectors are formed on the basis of measurements of the uplink signal.

4. The method as claimed in claim 1 , wherein the short-term covariance matrices ( R xx ) i are each obtained from a measurement, extending over one time slot of the downlink signal.

5. The method as claimed in claim 4 , wherein the first spatial covariance matrix ( R xx ) z , which is obtained by averaging the short-term covariance matrices, is a long-term covariance matrix.

6. The method as claimed in claim 1 , wherein

eigenvalues are obtained for the eigenvectors of the first spatial covariance, and

the eigenvectors of the first spatial covariance matrix, which have the greatest, are selected as first weighting vectors.

7. The method as claimed in claim 1 , further comprising:

providing short-term covariance matrices for each tap of the downlink signal;

producing a first spatial covariance matrix of the downlink signal by averaging the short-term covariance matrices over all taps;

determining eigenvectors of the first spatial covariance matrix;

cyclically regenerating the short-term covariance matrices to produce regenerated short-term covariance matrices;

modifying the eigenvectors of the first spatial covariance matrix using the regenerated short-term covariance matrices, to thereby produce modified eigenvectors;

calculating eigenvalues for the modified eigenvectors; and

selecting the first weighting vectors from among the modified eigenvectors, based on the eigenvalues of the regenerated short-term covariance matrices.

8. The method as claimed in claim 7 , wherein the eigenvectors of the first spatial covariance matrix are modified by multiplying the eigenvectors of the first spatial covariance matrix with the regenerated short-term covariance matrices.

9. The method as claimed in claim 1 , wherein the current weighting vector is selected from among the first weighting vectors.

10. The method as claimed in claim 1 , wherein the current weighting vector is a linear combination of the first weighting vectors.

11. The method as claimed in claim 1 , wherein training sequences are transmitted during the initialization phase.

12. The method as claimed in claim 1 , wherein the initialization phase is repeatedly performed.

13. The method as claimed in claim 1 , further comprising:

producing a first spatial covariance matrix for an uplink signal transmitted from the subscriber stations to the base station,

determining eigenvectors of the first spatial covariance matrix; and

selecting the first weighting vectors from among the eigenvectors of the first spatial covariance matrix.

14. The method as claimed in claim 13 , wherein

short-term covariance matrices are formed for each tap of the uplink signal; and

the first spatial covariance matrix is obtained by averaging the short-term covariance matrices over all taps.

15. The method as claimed in claim 14 , wherein the short-term covariance matrices ( R xx ) i are each obtained from a measurement, extending over one time slot of the uplink signal.

16. The method as claimed in claim 15 , wherein the first spatial covariance matrix ( R xx ) z , which is obtained by averaging the short-term covariance matrices, is a long-term covariance matrix.

17. The method as claimed in claim 15 , wherein

eigenvalues are obtained for the eigenvectors of the first spatial covariance, and

the eigenvectors of the first spatial covariance matrix, which have the greatest, are selected as first weighting vectors.

18. The method as claimed in claim 15 , wherein

cyclically regenerating the short-term covariance matrices to produce regenerated short-term covariance matrices;

modifying the eigenvectors of the first spatial covariance matrix using the regenerated short-term covariance matrices, to thereby produce modified eigenvectors;

calculating eigenvalues for the modified eigenvectors; and

selecting the first weighting vectors from among the modified eigenvectors, based on the eigenvalues of the regenerated short-term covariance matrices.

19. The method as claimed in claim 18 , wherein the eigenvectors of the first spatial covariance matrix are modified by multiplying the eigenvectors of the first spatial covariance matrix with the regenerated short-term covariance matrices.

Assignments (9)