An intelligent eNodeB for lte uplink based on neural network

The use of Levenberg Marquardt (LM) and gradient descent (GD) based machine learning algorithms in feed forward random neural network (RNN) for preparing calculation based structure is utilized to enhance radio resource management (RRM) and inter cell interference coordination (ICIC) in LTE framework. In this paper neural system dependent cognitive engine is implanted inside eNodeB which coordinately propose ideal radio factor to the clients, and optimum transmit energy to the working clients by neighboring cells. Less computational complexity, quick decision making, and long term learning are fundamental necessities to outline to convey methodically in any cognitive communication framework and the greater part of the existing strategies utilized as a cognitive method need in. The phenomenon of feed forward RNN upheld structure is inspected with conventional methods. To guarantee a superior execution of the framework, the outcomes are checked and contrasted with conventional methods.