System and method for selecting transmission parameters

1. A computer implemented method for selecting transmission (TX) parameters, the method comprising:

a. receiving a wireless signal, the wireless signal comprising reference signals;

b. generating a signal model for rank=2, based on reference signals of a received wireless signal;

c. converting, using QR decomposition, the signal model to a four-parameter representation, wherein the QR decomposition comprises decomposing the signal model into matrices Q and R, wherein matrix Q is unitary and matrix R is an upper triangular matrix, wherein the four-parameter representation is derived from the upper triangular matrix R of the QR decomposition;

d. determining, based on the four-parameter representation, whether mutual information per bit (MIB) values depend on a single parameter or on a plurality of parameters;

e. if the MIB values depend on the single parameter, calculating MIB values based on the single parameter;

f. if the MIB values depend on the plurality of parameters, calculating MIB values based on the plurality of parameters; and

g. selecting TX parameters that provide a maximal expected throughput based on the MIB values.

2. The method of claim 1 , wherein calculating MIB values based on the single parameter is performed using a machine learning algorithm.

3. The method of claim 1 , wherein operations d and f are performed using a machine learning algorithm.

4. The method of claim 3 , wherein the machine learning algorithm is an artificial neural network.

5. The method of claim 1 , wherein operation f is performed using an artificial neural network, and wherein the artificial neural network provides a plurality of the MIB values in parallel.

6. The method of claim 1 , further comprising:

h. repeating operations d-f for a plurality of channel quality indications (CQI);

i. repeating operations b-f for a plurality of pre-coder matrix indications (PMI);

j. estimating required block error rate (BLER) values for combinations of PMI and CQI for rank=2 based on the calculated MIB values;

k. estimating BLER values for combinations of PMI and CQI for rank=1; and

l. selecting a combination of RI, PMI and CQI that provides a maximal expected throughput for which BLER is not greater than a predetermined value.

7. The method of claim 6 , further comprising:

m. reporting the selected combination of RI, PMI and CQI to the base station; and

n. receiving data transmitted from the base station using the selected combination of RI, PMI and CQI.

8. The method of claim 1 , wherein the wireless signal is transmitted over a multiple-input-multiple-output (MIMO) channel.

9. A computer implemented method for selecting transmission (TX) parameters, the method comprising:

a. generating a signal model for rank=2, based on reference signals of a received wireless signal;

b. converting, by QR decomposition of an effective channel matrix, the signal model to a four-parameter representation, wherein the QR decomposition comprises decomposing the signal model into matrices Q and R, wherein matrix Q is unitary and matrix R is an upper triangular matrix, wherein the four-parameter representation is derived from the upper triangular matrix R of the QR decomposition; and

c. calculating mutual information per bit (MIB) values based on the four-parameter representation using an artificial neural network;

d. selecting TX parameters that provide a maximal expected throughput based on the MIB values.

10. The method of claim 9 , comprising:

e. repeating operation c for a plurality of CQIs;

f. repeating operations a-c for a plurality of PMIs;

g. estimating required BLER values for combinations of PMI and CQI for rank=2 based on the calculated MIB values;

h. estimating block error rate (BLER) values for combinations of pre-coder matrix indications (PMI) and channel quality indication (COI) for rank=1;

i. selecting a combination of RI, CQI and PMI that provides a maximal expected throughput for which BLER is not greater than a predetermined value;

j. reporting the selected combination of RI, CQI and PMI to the base station; and

k. receiving data transmitted from the base station using the selected combination of RI, CQI and PMI.

11. A device for selecting transmission (TX) parameters, the device comprising:

a memory;

a processor configured to:

a. generate a signal model for rank=2, based on reference signals of a received wireless signal;

b. convert, using QR decomposition, the signal model to a four-parameter representation, wherein the QR decomposition comprises decomposing the signal model into matrices Q and R, wherein matrix Q is unitary and matrix R is an upper triangular matrix, wherein the four-parameter representation is derived from the upper triangular matrix R of the QR decomposition;

c. determine, based on the four-parameter representation, whether mutual information per bit (MIB) values depend on a single parameter or on a plurality of parameters;

d. if the MIB values depend on the single parameter, calculate MIB values based on the single parameter; and

e. if the MIB values depend on the plurality of parameters, calculate MIB values based on the plurality of parameters;

f. selecting TX parameters that provide a maximal expected throughput based on the MIB values.

12. The device of claim 11 , wherein the processor is configured to calculate MIB values based on the single parameter using a machine learning algorithm.

13. The device of claim 11 , wherein the processor is configured to perform operations c and e using a machine learning algorithm.

14. The device of claim 13 , wherein the machine learning algorithm is an artificial neural network.

15. The device of claim 11 , wherein the processor is configured to perform operation e using an artificial neural network, and wherein the artificial neural network provides a plurality of the MIB values in parallel.

16. The device of claim 11 , wherein the processor is further configured to:

g. repeat operations c-e for a plurality of channel quality indications (CQI);

h. repeat operations a-e for a plurality of pre-coder matrix indications (PMI);

i. estimate required block error rate (BLER) values for combinations of PMI and CQI for rank=2 based on the calculated MIB values;

j. estimate BLER values for combinations of pre-coder matrix indications and CQI for rank=1; and

k. select a combination of RI, CQI and PMI that provides a maximal expected throughput for which BLER is not greater than a predetermined value.

17. The device of claim 16 , wherein the processor is further configured to:

l. report the selected combination of RI, CQI and PMI to the base station; and

m. receive data transmitted from the base station using the selected combination of RI, CQI and PMI.

18. The device of claim 11 , wherein the wireless signal is transmitted over a multiple-input-multiple-output (MIMO) channel.