A Fuzzy-Neural Network Model for Balancing Loads on a Wireless Network

Abstract

Over the years, there has been a growing demand for mobile communications which has become essential in the day to day activities of  humans. This calls for a robust and dynamic networks that with stand the rigour of the ever increasing demand for mobile  communication. Radio frequency channels have to be utilized as much as possible and are a precious resource. Various techniques have  been used to enhance the efficiency and flexibility of mobile communication networks to deal with the need of increased traffic and user  needs and characteristics. Load balancing is a critical selforganizing operation that ensures an equitable distribution of network users.  Several channel allocation schemes, including fixed channel assignment (FCA) and dynamic channel allocation (DCA) were proposed to  assign frequencies to cells in order to maximize frequency resumption (HCA).This work presents Fuzzy-Neural Network model in the  balancing of loads on a wireless network. Firstly, fuzzy rules are applied at level 1. Secondly, an adaptive fuzzy-neural network is designed and simulated. Adaptive Neuro-Fuzzy Inference System (ANFIS) is enhanced by expert knowledge of the fuzzy deduction method and  neural network capacities.