Low-complexity precoder design for large-scale MIMO communication systems

摘要

A method and system are provided. The method includes providing transmit precoders for a multiple-input and multiple-output communication system having a plurality of transmit antennas. The plurality of transmit antennas are for forming, using precoding, a plurality of channels such that each of the plurality of channels are configurable to serve a respective one of a plurality of users. The providing step includes imposing a respective average transmit antenna power constraint on each of the plurality of transmit antennas. The providing step further includes determining a diagonal precoder responsive to applying column scaling to a downlink channel matrix having a plurality of rows and a plurality of columns. The providing step additionally includes generating, from the diagonal precoder, a weighted precoder in accordance with the respective average antenna power constraint by optimizing a weighted sum-rate obtained upon transmitting respective signals over the plurality of channels.

主权项

1. A method, comprising:
providing transmit precoders for a multiple-input and multiple-output communication system having a plurality of transmit antennas, the plurality of transmit antennas for forming, using precoding, a plurality of channels such that each of the plurality of channels are configurable to serve a respective one of a plurality of users, wherein said providing step comprises:
imposing a respective average transmit antenna power constraint on each of the plurality of transmit antennas;determining a diagonal precoder responsive to applying column scaling to a downlink channel matrix having a plurality of rows and a plurality of columns; andgenerating, from the diagonal precoder, a weighted precoder in accordance with the respective average antenna power constraint by:
optimizing a weighted sum-rate obtained upon transmitting respective signals over the plurality of channels; andperforming an iterative stochastic power allocation pairwise bisection search, wherein the iterative stochastic power allocation pairwise bisection search includes, for each of a plurality of iterations, randomly selecting two users from among the plurality of users and locally allocating power between the two users to maximize an objective function for determining the optimal weighted sum-rate.