A Multi-source Direction Dimension-degraded Estimation Approach Based on Steering-vector
Pengfei Li · Journal of Projectiles.Rockets.Missiles and Guidance · 2011
To solve the modeling problem of multi-source direction estimation based on smart learning method,an effective dimension-degraded modeling approach was proposed in this paper.This approach based on decomposing the DOA matrix could result in the steering-vector of each signal source,then they were taken as the input vector to train the radial basis function neural networks(RBFNN) model with the training sets of single source,the trained model could approximate the nonlinear mapping from the separating input vectors of multi-source to direction-of-arrival.The simulation results showed the effectiveness and feasibility of this method.