Evolving Radial Basis Function Neural Network with Hausdorff Similarity Measure for SONAR signals detection/ classification

In this paper, a new approach has been proposed for detection/ classification of SONAR signals based on radial basis function neural network (RBFNN), which has been modified with a robust and reliable measure named: Hausdorff similarity measure (HSM). Methodologies of approach and simulation results are also represented. The final results show the new approach is able to increase the total performance of detection/ classification of SONAR targets even in low SNR.