Channel quality prediction using neural networks

Complex problems of wireless communications require a new, intelligent infrastructure having the capability of learning from measured traffic, adaptability and real time reaction. The properties of neural networks (NN) recommend them as an appropriate tool for telecommunications industry. This paper presents a channel quality prediction in a Long Term Evolution environment by using NN. Different NN architectures were tested and the best performance experiments results are presented. The results are promising, opening new possibilities to NN applications.