MAXIMIZING EFFICIENCY OF MULTI-USER COMMUNICATIONS NETWORKS

摘要

Systems and methods are described for jointly optimizing communications in a multi-user network to maximize the weighted sum-rate of its data links. For example, embodiments seek to maximize overall link transmit rates, while minimizing link interference and/or noise in general multi-in multi-out (MIMO) interference networks having multiple interfering data links, each with one or more antennas at the transmitter and receiver. Novel algorithmic approaches, including an “iterative minimax” approach and a “dual link” approach, are described for solving the weighted sum-rate maximization in a manner that converges quickly and reliably on an optimal point. Some embodiments implement these approaches in a centralized transmission controller, while others implement these approaches in a distributed fashion.

主权项

1. A method of link optimization in a multi-user communications network having a plurality of transmitter-receiver communications links that are mutually interfering, the method comprising:
initializing a forward-link signal communication model that represents transmit characteristics of a plurality of forward-link transmitters of the communications network; computing an initial solution to an optimization problem having at least one power constraint corresponding to the transmitters as a function of the initialized forward-link signal communication model; and converging on an optimal solution to the optimization problem by iterating, until an updated solution to the optimization problem satisfies a predetermined convergence criterion, steps comprising:
computing a forward-link interference model as a function of the forward-link signal communication model;updating a reverse-link signal communication model according to the forward-link interference model;computing a reverse-link interference model as a function of the updated reverse-link signal communication model;updating the forward-link signal communication model according to the reverse-link interference model;normalizing the updated forward-link signal communication model to fit the at least one power constraint; andcomputing the updated solution to the optimization problem as a function of the normalized updated forward-link signal communication model.