DOA estimation of near-field sources using support vector regression
N. Venkateswaran, C. Ashok · 2016
The problem of estimating the DOAs of multiple near field sources is resolved using SVR method. The proposed approach computes the DOAs from a nonlinear mapping function, which relates the received signal with its associated DOAs. The nonlinear mapping function is approximated by support vector regression (SVR) model. The proposed novel method yields better performance in comparison with the existing methods such as Generalized ESPRIT (GESPRIT), proposed for near field source localization.