Big Data Analytics For Prediction Of Mobile Users Movement Using Neural Networks

The rise of data warehouses and the rise of multi-media, social media and the Internet of Thingsrn(IoT) generate an increasing Volume of structured, semi-structured and unstructured data. Towardsrnthe investigation of these large Volumes of data, Big Data and data analytics have becomernemerging research fields attracting the attention of academia, industry and governments. Researchers,rnentrepreneurs, decision makers and problem solvers view ‘Big Data’ as an importantrntool used to revolutionize various industries and sectors, such as business, health-care, retail,rnresearch, education and public administration.rnIn this context, a general view on analysis of Big Data, especially in telecommunications industryrnis proposed. In order to allocate scarce resources efficiently the location of mobile usersrnshould be predicted. So, this work focuses on analysis of data for mobile users movement predictionrnin telecommunications network. The objective of this work is to process and analyzernobtained samples of OpenCelliD data by means of Neural Network and provide as accurate mobilernusers movement prediction as possible. More specifically, Modified Levenberg-Marquardtrn(LM) algorithm is presented as an effective algorithm for movement prediction. Obtained resultrnfrom prediction is optimized by iteration method designed for finding the best possible combinationrnof Neural Network parameters. Efficiency of mobile users movement prediction is verifiedrnby simulation in Matrix Laboratory (MATLAB). Simulation results show sufficient accuracyrnfor wide use of prediction for mobile networks optimization or services exploiting predictionrnof mobile users movement. Measured results fully reflect real solution for telecommunicationsrnindustry and can help to plan activities connected with mobile users movement in a given area.

Get Full Work

Report copyright infringement or plagiarism