A neural network approach for DOA estimation and tracking

Many signal subspace-based approaches have been proposed for determining the fixed direction of arrival (DOA) of plane waves impinging on an array of sensors. However, the computational burden of subspace-based algorithms makes them unsuitable for real-time processing of nonstationary signal parameters. We present an iterative procedure for DOA estimation and tracking, The complete procedure consists of, first, extracting the noise or signal subspace, by training the MCA or PCA algorithms, respectively. These algorithms contain only relatively simple operations and have self-organizing properties. Then, using the Newton algorithm, we get the estimated DOA. The performance on simulated data representing both constant and time-varying signals of the approach is presented.