An Algorithm of DOA Based on Fourth-order Cumulants
Cheng Gang, Cdma Division · Tance yu kongzhi xuebao · 2003
A novel algorithm of DOA eatimation based on fourth-order eumulants is presented. The suitable adjustment of the components of the cumulant matrices make it more applicable to DOA problems by applied directly to the well developed algorithms such as ESPRIT. By using the proposed cumulants matrices along with ESPRIT algorithm, the virtual sensors (twice the actual sensors) are added and the computational cost is reduced. As the algorithm is insensitive to additive Gaussian noise, the resolvability can be enhanced. Simulation experiments are included to validate the analytical result and to illustrate the potential performances of the proposed algorithm.