Classification of objects buried in inhomogeneous medium by artificial neural network using data obtained by impulse GPR with 1 Tx+ 4Rx antenna system

The detection of objects in random inhomogeneous medium by processing signals received by means of ground penetrating radar using artificial neural network is presented in the work. The simulation of impulse wave propagation through inhomogeneous medium and reflection from an object is carried out by FDTD method. The medium is a model of a soil with inclusions of random placement, size, and permittivity. The buried objects are the model of real antipersonnel mines. Four-element antenna system produces four different signals used after discretization as input data for artificial neural networks. The network is trained to recognize the object and its position for different models of random inhomogeneities of the medium. Its work is checked for the case of presence of Gaussian noise in received signals.