In this article, we applied Deep Learning on LTE-A uplink channel estimation\nsystem. The work involved creating of two SC-FDMA databases for\ntraining and for test, based on three types of channel propagation models.\nThe first section of this work consists of applying an Artificial Neural Network\nto estimate the channel of SC-FDMA link. Neural Network training is\nan iterative process which consists on adapting the values of its parameters:\nweights and bias. After training, the Neural Network was tested and implemented\non the receiver. The second section of this work deals with the same\nexperimentation but by using Deep Learning instead of classic Neural Networks.\nThe simulation results showed a strong improvement given by Deep\nlearning compared to classic method concerning the Bit Error Rate and the\nprocess speed. The third part of this work had been reserved to the complexity\nstudy. We had proved that Deep Learning gives better performance than\nMMSE estimator with low complexity.
Loading....