Network Traffic Optimization using k-NN Algorithm

Abstract

Traditional network optimization technology passively alter the network configurations because of network's congestion ratio, drop-off price, protection holes and so on., main to sub-ultimate user studies. As a result, the target is obtaining the proper network status, user requires and utility request distribution based on the real time data to optimize the network configurations. The data mining technique is introduced to expect the useful resource margin based totally on historic measurement facts. To discover the dynamic distribution of person demand and utility request, a weighted k-Nearest Neighbours (k-NN) model is proposed to calculate periodic characteristics of network traffics, denote temporal and spatial patterns of radio resource limits. Not like the conventional passive network optimization methods, the radio assets can be reconfigured actively to satisfy the dynamic pattern of traffic loads with the help of the usage of the proposed radio resource optimization using k-NN algorithm.