Method and network node for calculating transmitter precoding weights and receiver combining weights for a MIMO antenna system

1. A method in a network node implementing a multiple input multiple output (MIMO) antenna system, the method comprising:

estimating, at the network node, channel responses for user terminals accessing the network node on a carrier;

determining zero forcing beamforming weights for the carrier by adding one of the user terminals at a time in a calculation of an inverse of a Gram matrix containing parameters of the channel responses;

receiving, at the network node, downlink symbols for transmission towards one of the user terminals;

precoding the downlink symbols using the zero forcing beamforming weights; and

transmitting, from the network node on the carrier, the precoded downlink symbols towards the one of the user terminals via an array of M antennas.

2. The method of claim 1 , wherein the network node estimates the channel response for a given user terminal based on a predefined pilot signal received from the given user terminal.

3. The method of claim 1 , wherein the network node estimates directly or indirectly the channel response for a given user terminal based on a feedback received from the given user terminal.

4. The method of claim 1 , further comprising:

estimating, at the network node, a channel response for a further user terminal accessing the network node on the carrier; and

recalculating the inverse of the Gram matrix as a function of the channel response of the further user terminal;

whereby a number of columns and a number of rows of the Gram matrix are each incremented by one.

5. The method of claim 1 , further comprising:

detecting that a first one of the user terminals previously accessing the network node on the carrier has become disconnected from the network node;

permuting the previously calculated inverse of the Gram matrix so that a column and a row of the Gram matrix corresponding to the first one of the user terminals are placed on a last column and a last row of the Gram matrix; and

deleting the last column and the last row of the Gram matrix.

6. The method of claim 1 , further comprising:

detecting, at the network node, a change of a channel response for a second one of the user terminals accessing the network node on the carrier;

permuting the previously calculated inverse of the Gram matrix so that a column and a row of the Gram matrix corresponding to the second one of the user terminals are placed on a last column and a last row of the Gram matrix;

deleting the last column and the last row of the Gram matrix; and

recalculating the inverse of the Gram matrix as a function of the changed channel response of the second one of the user terminals;

whereby a number of columns and a number of rows of the Gram matrix are each first decremented by one and then incremented by one.

7. The method of claim 1 , wherein the carrier is a subcarrier of a multicarrier transmission system operating on a plurality of subcarriers, the method further comprising:

determining zero forcing beamforming weights for a subset of the plurality of subcarriers; and

calculating zero forcing beamforming weights for a remainder of the plurality of subcarriers by interpolating the zero forcing beamforming weights determined for the subset of the plurality of subcarriers.

8. The method of claim 1 , further comprising combining the calculation of the inverse of the Gram matrix with a polynomial expansion to resolve a generalized interference case.

9. The method of claim 1 , further comprising truncating a number of coefficients in the calculation of the inverse of the Gram matrix.

10. The method of claim 1 , further comprising estimating, at the network node, specific channel responses for particular user terminals accessing the network node at an edge of its coverage area, the specific channel responses forming a matrix projected on a subspace orthogonal of an interference channel internal to the coverage area of the network node.

11. The method of claim 1 , further comprising:

estimating, at the network node, specific channel responses of interfering user terminals located outside of a coverage area of the network node; and

recalculating the inverse of the Gram matrix as a function of the specific channel response of the interfering user terminals.

12. The method of claim 1 , wherein:

at least one of the user terminals is a multi-antenna user terminal; and

determining zero forcing beamforming weights for the carrier comprises independently adding each antenna of the multi-antenna user terminal in the calculation of the inverse of the Gram matrix.

13. A network node implementing a multiple input multiple output (MIMO) antenna system, the network node comprising:

an array of M antennas adapted to transmit signals toward user terminals accessing the network node on a carrier and to receive signals from the user terminals;

an estimator of channel responses received on the array of M antennas from the user terminals;

a weight calculator of zero forcing beamforming weights for the carrier by adding one of the user terminals at a time in a calculation of an inverse of a Gram matrix containing parameters of the channel responses;

a precoding module adapted to receive downlink symbols for transmission towards one of the user terminals and to precode the downlink symbols using the zero forcing beamforming weights, wherein the precoding module is implemented using one of software, firmware, hardware and a combination thereof; and

a transmitter adapted to transmit the precoded downlink symbols towards the one of the user terminals on the carrier via the array of M antennas.

14. The network node of claim 13 , further comprising:

at least one processor; and

a memory coupled to the processor and comprising non-transitory code instructions that when executed cause the processor to implement the estimator and the weight calculator.

15. The network node of claim 13 , wherein the channel responses depend at least in part on the number M of antennas.

16. The network node of claim 13 , further comprising:

an inverse Fast Fourier Transform (IFFT) module adapted to process the precoded downlink symbols before their transmission via the array of M antennas.

17. The network node of claim 13 , further comprising:

a receiver of uplink symbols from a fourth one of the user terminals on the carrier;

a detection module adapted to use the zero forcing beamforming weights to combine the uplink symbols; and

a Fast Fourier Transform (FFT) module adapted to process the received uplink symbols before their detection by the detection module.

18. The network node of claim 13 , wherein network node is adapted to communicate with the user terminals using a multicarrier transmission system operating on a plurality of subcarriers.

19. The network node of claim 13 , wherein:

the estimator is adapted to estimate a distinct channel response for each antenna of a multi-antenna user terminal; and

the weight calculator is adapted to calculate zero forcing beamforming weights for the carrier by independently adding each antenna of the multi-antenna user terminal in the calculation of the inverse of the Gram matrix.