URSI 2003, XVIII Simposium Nacional de la Unión Científica Internacional de Radio, La Coruña
Blind beamforming is a common problem in wireless
communications, where an array of antennas receives a
number of signals from distinct locations at the same frequency
and at the same time. In this paper the problem of blind beamforming
for multiple constant modulus (CM) signals separation
is solved using support vector machine (SVM) techniques. The
CM property of the signal is used to formulate a regression
problem which can be adapted to the SVM scheme, leading to
an iterative reweighted algorithm. Once a signal is recovered,
its contribution to the original observations is removed and the
iterative procedure can be applied again to extract another CM
signal. Simulation results show that this SVM-based algorithm
offers better performance than the algebraic constant modulus
algorithm (ACMA), mainly when only a small number of snapshots
is available.
