Scholarlink

In the realm of 5G and beyond 5G (B5G) wireless communication systems, various strategies have been proposed to optimize power allocation (PA). Heterogeneous networks (HetNets), a key component of 5G, have introduced new concepts such as ON-OFF switching of the base station (BS), user association and energy harvesting (EH) to improve energy efficiency (EE) and PA. This paper presents a novel technique to enable a BS to switch between active and sleep modes to improve EE. The main goal is to minimize total power consumption while accommodating incoming user traffic seamlessly. The approach uses advanced artificial intelligence tools, specifically genetic algorithms (GAs) and particle swarm optimization (PSO), to develop a cell switch-off scheme. Simulations have shown that this scheme improves scalability and efficiency. The comparative analysis of the proposed methods shows promising results under four scenarios and results in higher EE (7.43 bit/J).

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

Dynamic ON-OFF Energy Efficiency in HetNets Based Genetic and Particle Swarm Optimization Techniques

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Khaled Hmidi,
  • Sameh Najeh,
  • Ammar Bouallegue

摘要

In the realm of 5G and beyond 5G (B5G) wireless communication systems, various strategies have been proposed to optimize power allocation (PA). Heterogeneous networks (HetNets), a key component of 5G, have introduced new concepts such as ON-OFF switching of the base station (BS), user association and energy harvesting (EH) to improve energy efficiency (EE) and PA. This paper presents a novel technique to enable a BS to switch between active and sleep modes to improve EE. The main goal is to minimize total power consumption while accommodating incoming user traffic seamlessly. The approach uses advanced artificial intelligence tools, specifically genetic algorithms (GAs) and particle swarm optimization (PSO), to develop a cell switch-off scheme. Simulations have shown that this scheme improves scalability and efficiency. The comparative analysis of the proposed methods shows promising results under four scenarios and results in higher EE (7.43 bit/J).