Two-stage precoding and user grouping for large scale multiple-input multiple-output (MIMO) systems
Methods are provided. A method includes determining respective weighted likelihoods corresponding to a plurality of users in a multiple-input multiple-output communication system. The method further includes forming a plurality of user groups from the plurality of users using an iterative K-means clustering technique applied to the plurality of users. The iterative K-means clustering technique is responsive to the respective weighted likelihoods.
1. A method, comprising:
determining respective weighted likelihoods corresponding to a plurality of users in a multiple-input multiple-output communication system;
forming a plurality of user groups from the plurality of users using an iterative K-means clustering technique applied to the plurality of users, the iterative K-means clustering technique being responsive to the respective weighted likelihoods; and
incorporating the plurality of user groups into the multiple-input multiple-output communication system to increase throughput and load balancing in the multiple-input multiple-output communication system,
wherein respective group centers of the plurality of groups are determined using Eigen-decomposition.
2. The method of claim 1 , wherein the respective weighted likelihoods are determined responsive to a respective covariance matrix to each or a plurality of users and the respective group centers of the plurality of groups.
3. The method of claim 1 , wherein the respective weighted likelihoods are determined responsive to a plurality of different Eigenmodes.
4. The method of claim 3 , wherein the respective group centers and total likelihoods for each of the plurality of groups are determined responsive to the plurality of different Eigenmodes.
5. The method of claim 3 , wherein the plurality of different Eigenmodes is determined responsive to a respective covariance matrix determined for each of the plurality of users.
6. The method of claim 1 , further comprising approximating respective signal-to-interference-plus-noise ratios for each of the plurality of users.
7. The method of claim 6 , wherein the respective signal-to-interference-plus-noise ratios are determined responsive to user grouping information and user scheduling information for a given time slot.
8. The method of claim 6 , further comprising:
finding a maximum signal-to-interference-plus-noise ratio from among the respective signal-to-interference-plus-noise ratios; and
scheduling a given one of the plurality of users with the maximum signal-to-interference-plus-noise ratio for the given time slot.
9. The method of claim 6 , further comprising determining a sum rate of at least some of the plurality of users responsive to corresponding ones of the respective signal-to-interference-plus-noise ratios determined therefor.
10. The method of claim 6 , wherein the respective signal-to-interference-plus-noise ratios are determined responsive to respective pre-beamforming matrices determined for each of the plurality of groups.
11. The method of claim 6 , wherein each of the respective signal-to-interference-plus-noise ratios is determined responsive to a respective precoding matrix.
12. The method of claim 11 , wherein the respective precoding matrix corresponds to a two-stage precoding technique wherein interferences between the plurality of groups are suppressed in a first stage and interferences within each of the plurality of groups are suppressed in a second stage.
13. The method of claim 1 , wherein said forming step forms the plurality of groups using a load balancing technique.
14. A non-transitory article of manufacture tangibly embodying a computer readable program which when executed causes a computer to perform the steps of claim 1 .