A Handover Decision Optimization Method Based on Data-Driven MLP in 5G Ultra-Dense Small Cell HetNets
Küçük Resim Yok
Tarih
2025
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Springer
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
Deploying Ultra-Dense Small Cells (UDSCs) in Heterogeneous Networks (HetNets) introduces advantages such as increased capacity and expanded coverage over conventional HetNets. However, these advantages come at the expense of some challenges during the Handover (HO) process. Radio Link Failure (RLF) and Unnecessary Handover (UHO) are severe among these challenges. To address these issues, accurate setting and optimization of Handover Control Parameters (HCPs), including Handover Margin (HOM) and Time-To-Trigger (TTT), are necessary. Inaccurate adjustment and optimization of HCPs in live networks may lead to underperformance. Thus, this paper proposes a method that optimizes the obtained dataset by developing an algorithm that adjusts HOM and TTT based on related metrics such as RLF and UHO. The optimized dataset is then applied to a Multi-Layer Perception (MLP) model within a developed HO decision algorithm to predict both HOM and TTT, considering user speed, Reference Signal Received Power (RSRP), Signal to Interference plus Noise Ratio (SINR), and cell load. Simulation results showed that the proposed method outperforms the well-known A3 method in terms of Handover Rate (HOR), Handover Failure (HOF), Handover Ping-Pong (HOPP) and RLF by approximately 90.9%, 76.6%, 79.8% and 75%, respectively.
Açıklama
Anahtar Kelimeler
HetNets, Handover, HCPs, LTE-A, Machine learning, Ultra-dense small cell networks, 5G
Kaynak
Journal of Networkand Systems Management
WoS Q Değeri
Q2
Scopus Q Değeri
Q1
Cilt
33
Sayı
2
