Granted Patent B2

US 7,123,942 · App. 10/200,627 · Granted Oct 17, 2006

Abstract

Claims (20)

1. A method for controlling transmission from a first radio station having N transmission antennas to a second radio station, N being a number larger than 2, comprising the following steps:

broadcasting M beacon signals from the first station, M being a number larger than 1 and smaller than N, each beacon signal being associated with a distribution vector with N components and being applied to each of the N antennas after weighting by a corresponding component of the associated distributor vector;

evaluating the reception of the M beacon signals by the second station, so as to determine feedback data communicated to the first station;

causing the distribution vectors associated with the M beacon signals to vary over time;

processing the feedback data successively received by the first station so as to determine a respective weighting coefficient for each of the N antennas; and

transmitting a dedicated radio signal from the first station destined for the second station, the said dedicated signal being applied to each of the N transmission antennas after weighting by the coefficient determined for the said antenna.

2. A method as claimed in claim 1 , wherein the processing of the feedback data comprises an estimation of a matrix with N rows and N columns on the basis of feedback data obtained for several combinations of distribution vectors, the term situated in the (i+1)-th and in the (j+1)-th column of the said matrix, for 0≦i<N and 0≦j<N, being proportional to h′ j h .h′ i , where h′ i denotes an impulse response of a propagation channel between the (i+1)-th transmission antenna of the first station and the second station, and (.) H denotes the conjugate transpose.

3. A method as claimed in claim 2 , wherein the processing of the feedback data comprises the calculation of an eigenvector corresponding to an eigenvalue of maximum modulus of the said matrix with N rows and N columns, the respective weighting coefficients for the N antennas being proportional to the components of the said eigenvector.

4. A method as claimed in claim 2 , wherein the said matrix with N rows and N columns is estimated on the basis of the feedback data obtained for (N−1)(N+2)/2 distinct successive combinations of distribution vectors.

5. A method as claimed in claim 2 , wherein the estimation of the said matrix with N rows and N columns comprises an updating of the said matrix by interpolation on the basis of the feedback data obtained for each combination of distribution vectors.

6. A method as claimed in claim 1 , wherein each distribution vector associated with a beacon signal has a single nonzero component.

7. Radio station with closed-loop transmission diversity, comprising N transmission antennas, N being a number larger than 1, and comprising

means for broadcasting M beacon signals, M being a number larger than 1, said means for broadcasting comprising means for associating each of the M beacon signals with a respective distribution vector with N components, and means for applying each of the M beacon signals to each of the N antennas after weighting by a corresponding component of the associated distribution vector,

means for transmitting at least one dedicated radio signal destined for a respective remote radio station, said means for transmitting the dedicated radio signal destined for a remote radio station comprising means for processing feedback data received from the said remote station so as to determine a respective weighting coefficient for each of the N antennas, and means for applying the said dedicated signal to each of the N transmission antennas after weighting by the coefficient determined for the said antenna,

wherein N>M, wherein the broadcasting means furthermore comprise means for causing the distribution vectors associated with the M beacon signals to vary over time, and wherein the processing means are devised so as to determine the respective weighting coefficients for the N antennas on the basis of feedback data successively received for different combinations of distribution vectors.

8. Radio station as claimed in claim 7 , wherein the means for processing the feedback data comprise means for estimating a matrix with N rows and N columns on the basis of feedback data obtained for several combinations of distribution vectors, the term situated in the (i+1)-th and in the (j+1)-th column of the said matrix, for 0≦i<N and 0≦j<N, being proportional to h′ j H .h′ i , where h′ i denotes an impulse response of a propagation channel between the (i+1)-th transmission antenna of the first station and the second station, and (.) H denotes the conjugate transpose.

9. Radio station as claimed in claim 8 , wherein the means for processing the feedback data comprise means for calculating an eigenvector corresponding to an eigenvalue of maximum modulus of the said matrix with N rows and N columns, the respective weighting coefficients for the N antennas being proportional to the components of the said eigenvector.

10. Radio station as claimed in claim 8 , wherein the means for estimating the matrix with N rows and N columns comprise means of interpolation for updating of the said matrix on the basis of the feedback data obtained for each combination of distribution vectors.

11. Radio station as claimed in claim 7 , wherein each distribution vector associated with a beacon signal has a single nonzero component.

12. Radio station as claimed in claim 7 , wherein N=4 and M=2.

Assignments (9)