Scholarlink

<p>In this paper, we propose an Automatic Handover Decision Model for 5th Generation Networks based on Age of Information (AoI) using Machine learning techniques. In this model, the default handover choice is overridden by applying the XGBoost classifier, based on the user’s historical handover success rates. The users in this network model are categorized into the 3 groups based on the type of service provided. The average AoI is estimated based on destination times, generation times and arrival rate. The AoI is analysed in terms of Signal-to-Noise Ratio at each receiving node and Block Error Rate at each destination. The proposed model performance is assessed using connection level Quality of Service (QoS) metrics, specifically the call blocking probability and handoff call dropping probability. Simulations are conducted to determine the effects of various metrics variations on the QoS metrics, including capacity, call arrival rate, departure rate, and required bandwidth unit.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Automatic handover decision model for 5G networks based on age of information using machine learning techniques

  • 期刊论文

  • 发表日期：

    2025年12月27日

查看全文
  • SRIDEVI S,
  • Jacob Augustine

摘要

In this paper, we propose an Automatic Handover Decision Model for 5th Generation Networks based on Age of Information (AoI) using Machine learning techniques. In this model, the default handover choice is overridden by applying the XGBoost classifier, based on the user’s historical handover success rates. The users in this network model are categorized into the 3 groups based on the type of service provided. The average AoI is estimated based on destination times, generation times and arrival rate. The AoI is analysed in terms of Signal-to-Noise Ratio at each receiving node and Block Error Rate at each destination. The proposed model performance is assessed using connection level Quality of Service (QoS) metrics, specifically the call blocking probability and handoff call dropping probability. Simulations are conducted to determine the effects of various metrics variations on the QoS metrics, including capacity, call arrival rate, departure rate, and required bandwidth unit.