In this project, we are aiming to study one of the most advanced mobile cellular telecommunication technology, which are going to be widely spread among users in the coming years, focusing on self-organizing network in Long Term Evolution (LTE) by using neural network technique.
LTE is designed to have wider channels up to 20MHz, with low latency less than 10 ms and packet optimized OFDMA radio access technology. The peak data rate for LTE is 100Mbps in downlink and 50 Mbps in the uplink without using MIMO technology.
This project is related to studying the scheduler types and its throughput, that is provided by LTE system and studying the neural network properties and its models. All these previous technologies are going to be tested by written MATLAB code.
This valuable jewel of this project is that, we were able to write a research paper that shows how to intelligently train base station (E-NodeB) to choose the most appropriate and optimized scheduler in LTE system for each pixel inside an image using Neural Network technique leading to save, utilize bandwidth & increase capacity, that is because neural network changes scheduler assigning process from fixed to dynamic in time based upon the importance of data that is being sent.
Keywords: LTE; NN; Scheduler; image pixel, E-NodeB