DOA estimation based on support vector machine ensemble

Abstract

Abstract A synthesis model based on support vector machine ensemble (SVME) is proposed for the direction of arrival (DOA) estimation. This method is based on the advantages of the support vector machine (SVM), such as high fitting precision, simple structure, and strong generalization ability. The basic concept of the method is to select SVMs, optimally, to construct an SVME with the aid of the binary particle‐swarm optimization algorithm. The performance of the proposed model is validated by comparing its simulation results with that of neural networks and a few other state‐of‐the‐art models. Experiments show that this method can reduce the prediction error and improve the generalization ability, using a limited number of training samples.