Title :
Optimal weights computation of an emitting antenna array - the Obele algorithm
Author :
Bertrand, Joël ; Forster, Philippe
Author_Institution :
Lab. Electronique et Commun., Conservatoire Nat. des Arts et Metiers, Paris, France
fDate :
7/1/2003 12:00:00 AM
Abstract :
The capacity of a mobile network can be increased by using antenna arrays. We present a new method that jointly optimizes antenna weights and powers for all users on the downlink on a single cell. The algorithm, which only requires the estimation of the noise power σ2 and knowledge of the channel responses on the downlink, consists of minimizing the greatest eigenvalue of a matrix. This optimization results from an iterative process. In a first pass, we compute the greatest eigenvalue of a matrix - the power is given by the associated eigenvector - which depends on the antenna weights, and in a second pass, we compute a new set of antenna weights that will be reused in the next iteration. We prove that it converges to the global minimum.
Keywords :
array signal processing; cellular radio; convergence of numerical methods; eigenvalues and eigenfunctions; matrix algebra; optimisation; radio links; radio networks; Obele algorithm; antenna array; antenna power optimization; antenna weights optimization; cellular radio network; channel responses; convergence; downlink; eigenvector; global minimum; iterative process; matrix eigenvalue; mobile network capacity; noise power estimation; optimal weights computation; power control; Antenna arrays; Base stations; Covariance matrix; Downlink; Eigenvalues and eigenfunctions; Interference; Mobile antennas; Optimization methods; Pulse modulation; Receiving antennas;
Journal_Title :
Signal Processing, IEEE Transactions on
DOI :
10.1109/TSP.2003.813582