2020/06/11 by Zoraze Ali, Ali, Zoraze, Marco Miozzo +9
Computer Science · Engineering · #Advanced Data and IoT Technologies #Caching and Content Delivery #FOS: Computer and information sciences #Internet Traffic Analysis and Secure E-voting #Machine Learning (cs.LG) #Networking and Internet Architecture (cs.NI)
paper · pdf · doi:10.48550/arxiv.2006.06526
openalex publication_date 2020/06/11 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
In this paper, we discuss a handover management scheme for Next Generation\nSelf-Organized Networks. We propose to extract experience from full protocol\nstack data, to make smart handover decisions in a multi-cell scenario, where\nusers move and are challenged by deep zones of an outage. Traditional handover\nschemes have the drawback of taking into account only the signal strength from\nthe serving, and the target cell, before the handover. However, we believe that\nthe expected Quality of Experience (QoE) resulting from the decision of target\ncell to handover to, should be the driving principle of the handover decision.\nIn particular, we propose two models based on multi-layer many-to-one LSTM\narchitecture, and a multi-layer LSTM AutoEncoder (AE) in conjunction with a\nMultiLayer Perceptron (MLP) neural network. We show that using experience\nextracted from data, we can improve the number of users finalizing the download\nby 18%, and we can reduce the time to download, with respect to a standard\nevent-based handover benchmark scheme. Moreover, for the sake of\ngeneralization, we test the LSTM Autoencoder in a different scenario, where it\nmaintains its performance improvements with a slight degradation, compared to\nthe original scenario.\n