Downlink beamforming method

• US 6,694,155 B1
• Filed: 09/05/2000
• Issued: 02/17/2004
• Est. Priority Date: 09/24/1999
• Status: Expired due to Fees

• Current Assignee: National University of Singapore
• Original Assignee: National University of Singapore
• Inventors: Liang, Ying Chang, Chin, Francois Po Shin
• Primary Examiner(s): Vuong, Quochien B.
• Application Number:
  US09/655,009
• Time in Patent Office: 1,260 Days
• Field of Search: 455/561, 455/562.1, 455/69, 455/63.1, 455/63.2, 455/446, 455/450
• US Class Current: 455/562.1
• CPC Class Codes: H04B 7/0615 of weighted versions of same signal
• CPC Class Codes: H04B 7/0617 for beam forming
• CPC Class Codes: H04L 2025/03426 transmission using multiple-input and multiple-output channels
• CPC Class Codes: H04L 25/0204 of multiple channels
• CPC Class Codes: H04L 25/021 Estimation of channel covariance
• CPC Class Codes: H04L 25/0248 Eigen-space methods
• CPC Class Codes: H04L 25/03343 Arrangements at the transmitter end

Abstract

A method for downlink capacity enhancement in a wireless communications system that has a base station with antenna array and terminals that are physically remote from the base station comprises the steps of receiving at the base station antenna array combinations of arriving signals from the plurality of remote terminals, estimating an uplink channel covariance matrix (UCCM) for each of the terminals from the combinations of arriving signals, constructing from each UCCM a downlink channel covariance matrix (DCCM), calculating from the DCCM a downlink weight vector for each of the terminals, transmitting a set of information signals from the base station antenna array according to the downlink weight vectors.

Claims

1. A method for downlink capacity enhancement in a wireless communications system comprising a base station with antenna array and terminals that are physically remote from said base station, the method comprising steps of:

receiving at said base station antenna array combinations of arriving signals from said plurality of remote terminals;

estimating an uplink channel covariance matrix (UCCM) for each of said terminals from said combinations of arriving signals;

constructing from each of said UCCM a downlink channel covariance matrix (DCCM);

inputting downlink data rate information (DRRI);

calculating from all said DCCM and DRRI a downlink weight vector for each of said terminals; and

transmitting a set of information signals from said base station antenna array according to said downlink weight vectors.

2. The method of claim 1 wherein the estimating step comprises:

forming from said combinations of arriving signals an uplink channel vector for each of said terminals;

establishing a UCCM for each said remote terminal by taking a linear combination of outer products of the corresponding the uplink channel vectors.

3. The method of claim 2 wherein the forming step comprises:

calculating from said combinations of arriving signals and sets of uplink training sequences associated with said remote terminals an uplink minimum mean-square-error (MMSE) weight vector for each said terminals;

assigning said uplink MMSE weight vector as the uplink channel vector for each of said terminals.

4. The method of claim 1 wherein the plurality of remote terminals are CDMA terminals, each of which has an unique PIN code sequence.

5. The method of claim 4 wherein the estimating step comprises:

forming a despread signal for each of said terminals from said combinations of arriving signals and associated PN code sequence; and

establishing a UCCM for each said remote terminal by taking a linear combination of outer products of the corresponding despread signal.

6. The method of claim 5 wherein the establishing step comprises:

computing an uplink channel vector for each of said terminals from the associated despread signal and at least one training sequence associated with each remote terminal; and

constructing a UCCM for each said remote terminal by taking a linear combination of outer products of the corresponding uplink channel vector.

7. The method of claim 6 wherein the computing step comprises:

calculating an estimated gradient of the error function that includes weighted magnitude square of said uplink channel vector

8. The method of claim 5 wherein the establishing step comprises:

computing an uplink minimum mean-square-error (MMSE) weight vector for each of said terminals from the associated despread signal; and

constructing a UCCM for each said remote terminal by taking a linear combination of outer products of the corresponding uplink weight vector.

9. The method of claim 8 wherein the computing step comprises:

calculating an estimated gradient of the error function that includes weighted magnitude square of said MMSE weight vector

10. The method of claim 1 wherein the constructing step comprises the substeps of:

columnising the said UCCM to a first column vector;

calculating a second column vector by multiplying a frequency calibration matrix M_A(FCM−

M_A) with said first column vector, the FCM-M_A, a n²×

m² matrix where m and n are the number of receive and transmit antenna elements, being only dependent on the carrier frequencies, transmit and receive array structures and cell sectorisation; and

constructing said DCCM from said second column vector.

11. The method of claim 1, wherein the UCCM is used as the DCCM in the constructing step.

12. The method of claim 1 wherein the constructing step comprises the substeps of:

extracting from the first column and first row of said UCCM to form a first column vector;

calculating a second column vector by multiplying a frequency calibration matrix M_B(FCM−

M_B) with said first column vector, the FCM−

M_B, a (n−

1)×

(2m−

1) matrix where m and n are the number of receive and transmit antenna elements, being only dependent on the carrier frequencies, transmit and receive array structures and cell sectorisation; and

constructing said DCCM from said second column vector.

13. The method of claim 1 wherein the constructing step comprises the substeps of:

extracting from the first column and first row of said UCCM to form a first column vector;

extracting the real part of said first column vector to form a second column vector and the imaginary part of said first column vector to form a third column vector;

calculating a fourth column vector by multiplying a frequency calibration matrix M_C(FCM−

M_C) with said second column vector, the FCM−

M_C, a n×

m matrix where m and n are the number of receive and transmit antenna elements, being only dependent on the carrier frequencies, transmit and receive array structures and cell sectorisation;

calculating a fifth column vector by multiplying a frequency calibration matrix M_D(FCM−

M_D) with said third column vector, the FCM−

M_D, a (n−

1)×

(m−

1) matrix where m and n are the number of receive and transmit antenna elements, being only dependent on the carrier frequencies, transmit and receive array structures and cell sectorisation;

forming a complex sixth column vector with real part being said fourth column vector and imaginary part being said fifth column vector; and

constructing said DCCM from said sixth column vector.

14. The method of claim 1 wherein the downlink weight vector for each of said terminals is the dominant eigenvector of the said DCCM corresponding to the said terminals.

15. The method of claim 1 wherein the calculating step comprises the substeps of:

calculating a channel vector for each of said terminals by taking the dominant eigenvector of the corresponding DCCM; and

repeating the steps of;

determining a set of power coefficients from a set of downlink system parameters concerning all mobile terminals that include said downlink weight vectors, said channel vector, downlink information data transmission rate and downlink link quality requirement of each of said terminals;

computing an auto correlation matrix by taking a weighted sum of all DCM corresponding to said terminals according to said set of power coefficients; and

forming a downlink weight vector for each of said terminals from said autocorrelation matrix and corresponding DCCM for said terminal, wherein the downlink weight vector has maximal projection onto the corresponding DCCM and minimal projection onto said autocorrelation matrix

until the said set of power coefficients and downlink weight vectors have converged.

16. The method of claim 15 wherein said downlink weight vector for each of said terminals is the dominant generalised eigenvector of the corresponding DCCM and said autocorrelation matrix.

17. The method of claim 15 wherein said downlink weight vector for each of said terminals is the dominant eigenvector of a matrix, which is the product of the inverse of said autocorrelation matrix and the corresponding DCCM.

18. The method of claim 15 wherein said downlink weight vector for each of said terminals is the product of the inverse of said autocorrelation matrix and the said corresponding channel vector.

19. The method of claim 1 wherein the calculating step comprises the substeps of:

calculating a channel vector for each of said terminals by taking the dominant eigenvector of the corresponding DCCM;

determining a set of power coefficients from a set of downlink system parameters concerning all mobile terminals that include said channel vector, downlink information data transmission rate and downlink link quality requirement of each of said terminals;

computing an autocorrelation matrix by taking a weighted sum of all DCCM corresponding to said terminals according to said set of power coefficients; and

forming a downlink weight vector for each of said terminals from said autocorrelation matrix and corresponding DCCM for said terminal, wherein the downlink weight vector has maximal projection onto the corresponding DCCM and minimal projection onto said autocorrelation matrix.

20. A method for downlink capacity enhancement in a wireless communications system comprising a base station with antenna array and terminals that are physically remote from said base station, the method comprising steps of:

receiving at said base station antenna array combinations of arriving signals from said plurality of remote terminals, wherein the plurality of remote terminals are CDMA terminals, each of which has an unique PN code sequence;

estimating an uplink weight vector for each of said terminals from said combinations of arriving signals including the steps of;

forming a despread signal for each of said terminals from said combinations of arriving signals and said associated PN code sequence; and

computing said uplink weight vector from corresponding despread signal;

constructing from each of said uplink weight vector a downlink weight vector;

transmitting the set of information signals from said base station antenna array according to said downlink weight vectors;

wherein the computing step comprises;

calculating a estimated gradient of the error function that includes weighted magnitude square of said uplink weight vector

21. A method for downlink capacity enhancement in a wireless communications system comprising a base station with antenna array and terminals that are physically remote from said base station, the method comprising steps of:

receiving at said base station antenna array combinations of arriving signals from said plurality of remote terminals;

estimating an uplink weight vector for each of said terminals from said combinations of arriving signals;

constructing from each of said uplink weight vector a downlink weight vector;

transmitting the set of information signals from said base station antenna array according to said downlink weight vectors; and

wherein the constructing step comprises;

determining the zeros of the polynomial whose coefficients are the elements of the uplink weight vector;

forming new polynomial zeros by scaling the phase of said zeros by a factor that is related to the ratio of the downlink frequency to the uplink frequency; and

establishing said downlink weight vector by constructing a new polynomial using said new polynomial zeros and using the coefficients of said new polynomial as the elements of said downlink weight vector.

22. A method for downlink capacity enhancement in a wireless communications system, comprising a base station with antenna array and terminals that at are physically remote from said base station, the method comprising steps of:

receiving at said base station antenna array combinations of arriving signals from said plurality of remote terminals;

estimating an uplink channel vector for each of said terminals from said combinations of arriving signals;

constructing from each of said uplink channel vector a downlink channel vector;

inputting downlink data rate information (DDRI);

calculating from all said downlink channel vector and DDRI a downlink weight vector for each of said terminals; and

transmitting the set of information signals from said base station antenna array according to said downlink weight vectors.

23. The method of claim 22 wherein the plurality of remote terminals are COMA terminals, each of which has an unique PN code sequence.

24. The method of claim 23 wherein the estimating step comprises:

forming a despread signal for each of said terminals from said combinations of arriving signals and said associated PN code sequence; and

computing said uplink channel vector from corresponding despread signal.

25. The method of claim 24 wherein the computing step comprises:

calculating an estimated gradient of the error function that includes weighted magnitude square of said uplink channel vector

26. The method of claim 24 wherein the computing step comprises:

establishing a UCCM for each said remote terminal by taking a linear combination of outer products of the corresponding despread signal; and

forming the uplink channel vector for each said remote terminal by taking the dominant eigenvector of corresponding said UCCM.

27. The method of claim 22 wherein the constructing step comprises:

determining the zeros of the polynomial whose coefficients are the elements of the uplink channel vector;

forming a new polynomial zeros by scaling the phase of said zeros by a factor that is related to the ratio of the downlink frequency to the uplink frequency; and

establishing said downlink channel vector by constructing a new polynomial using said new polynomial zeros and using coefficients of said new polynomial as the elements of said downlink channel vector.

28. The method of claim 22 wherein the uplink channel vector is used as the downlink channel vector in the constructing step.

29. The method of claim 22 wherein the downlink channel vector for each of said terminals is used as the corresponding downlink weight vector.

30. The method of claim 22 wherein the calculating step comprises repeating the substeps of:

determining a set of power coefficients from a set of downlink system parameters concerning all mobile terminals that include said downlink weight vectors, said downlink channel vector, downlink information data transmission rate and downlink link quality requirement of each of said terminals;

computing an autocorrelation matrix by taking a weighted sum of the outer products of said downlink channel vectors corresponding to said terminals according to said set of power coefficients and downlink information data transmission rates; and

forming a downlink weight vector for each of said terminals by taking the product of the inverse of said autocorrelation matrix and corresponding said downlink channel vector

until the said set of power coefficients and downlink weight vectors have converged.

31. The method of claim 22 wherein the calculating step comprises the substeps of:

determining a set of power coefficients from a set of downlink system parameters concerning all mobile terminals that include said downlink weight vectors, said downlink channel vector, downlink information data transmission rate and downlink link quality requirement of each of said terminals;

computing an autocorrelation matrix by taking a weighted sum of the outer products of said downlink channel vectors corresponding to said terminals according to said set of power coefficients and downlink information data transmission rates; and

forming a downlink weight vector for each of said terminals by taking the product of the inverse of said autocorrelation matrix and corresponding said downlink channel vector.

32. A base station for a wireless communications system, the base station comprising:

an uplink receive antenna array for receiving arriving signals from a plurality of remote terminals on respective uplink channels;

an uplink weight generator for estimating an uplink channel covariance matrix (UCCM) for each of said uplink channels;

a downlink weight generator operable to derive, from each of said UCCM, downlink weights, by constructing for each of said UCCM a downlink channel convergence matrix (DCCM), accepting inputted downlink data rate information (DDRI) and calculating from all DCCM and DDRI a downlink weight vector for each of said terminals; and

a downlink transmit antenna array to transmit signals to the remote terminals in accordance with the desired downlink weights.

33. A base station according to claim 32, wherein the uplink receive antenna array is the same as the downlink transmit antenna array.

34. A base station according to claim 32, wherein the uplink receive antenna array is separate from the downlink transmit antenna array.

35. A base station according to claim 32, wherein the property of the uplink channel comprises at least one of the uplink channel covariance matrices, the uplink channel responses or the uplink beamforming weights.

36. A base station according to claim 32, wherein uplink spatial- de-multiplexing means and downlink spatial multiplexing means are provided.

37. A communication system incorporating a base station according to claim 32 and plurality of remote terminals.