Compartmental Gaussian Mixture Model used for Layer Detection in Radio-Echo Sounding Data

This paper presents a compartmental Gaussian mixture model (GMM) method to trace layers of the ice sheet with radio-echo sounding (RES) data. Based on the compartmentalization of RES data, the proposed method build a compartmental Gaussian mixture model, which is solved using Fuzzy C-means (FCM) method and expectation maximization (EM) method to obtain preliminary layer detection results. And the layer boundaries of the ice sheet are detected according to analyzing the classification results of GMM. Experimental results show that the proposed method can trace layer boundaries of the ice sheet effectively.