Blind identification of FIR SIMO channels using an improved particle swarm optimizer

The blind multi-channels identification problem is studied in this paper. A cost function based on the orthogonal property between the output autocorrelation matrix and the channels parameter matrix is first constructed for a signal-input multiple-output FIR system. Then, an improved particle swarm optimizer, in which the personal best particle is replaced with the weight average of personal best particles in swarm and the acceleration coefficients is adaptively adjusted, is proposed to resolve the problem. Simulation results demonstrate that the improved particle swarm optimizer has good convergence performance, and is effective to resolve the blind FIR SIMO channels identification problem.