2021/11/27 by Muhammad Asif Khan, Ridha Hamila, Khan, Muhammad Asif +7 · 3 citations
Computer Science · Engineering · #Advanced MIMO Systems Optimization #Cooperative Communication and Network Coding #FOS: Computer and information sciences #Machine Learning (cs.LG) #Networking and Internet Architecture (cs.NI) #Wireless Networks and Protocols
paper · pdf · doi:10.48550/arxiv.2111.13879
openalex publication_date 2021/11/27 · openalex created_date 2022/11/17 · openalex updated_date 2026/07/28
Device mobility in dense Wi-Fi networks offers several challenges. Two\nwell-known problems related to device mobility are handover prediction and\naccess point selection. Due to the complex nature of the radio environment,\nanalytical models may not characterize the wireless channel, which makes the\nsolution of these problems very difficult. Recently, cognitive network\narchitectures using sophisticated learning techniques are increasingly being\napplied to such problems. In this paper, we propose data-driven machine\nlearning (ML) schemes to efficiently solve these problems in wireless LAN\n(WLAN) networks. The proposed schemes are evaluated and results are compared\nwith traditional approaches to the aforementioned problems. The results report\nsignificant improvement in network performance by applying the proposed\nschemes. The proposed scheme for handover prediction outperforms traditional\nmethods i.e. received signal strength method and traveling distance method by\nreducing the number of unnecessary handovers by 60% and 50% respectively.\nSimilarly, in AP selection, the proposed scheme outperforms the strongest\nsignal first and least loaded first algorithms by achieving higher throughput\ngains up to 9.2% and 8% respectively.\n