Enhancing Mobility Robustness and Load Balancing in Networks Using a Modified Deep CNN-BiLSTM Model with Attention Mechanism

Author:

Mohan Divya1,Mary A. Geetha1

Affiliation:

1. Vellore Institute of Technology

Abstract

Abstract

Mobile data transmission between nodes and well-known data services is always evolving, and these services offer a variety of communication endurance. Simultaneously, the 4G standards aim to address these services from the outset, recognizing that improving mobility resilience is necessary to achieve performance objectives in these challenging situations. Redefining new devices to meet the low latency and high-performance reliability requirements of mobile devices is essential. Ongoing advancements in mobile data transmission and services underscore the need for constant progress, emphasizing the crucial role of mobility robustness (MR) in optimizing handover-related parameters and enhancing user mobility performance. MR involves automated optimization in active and idle modes, ensuring optimal performance and end-user quality by considering attributes such as load balancing and neighbour relations. The handover procedure in the networking mechanism provides a transition to the base and the destination cell. This is followed by the load balancing (LB) technique, considered vital in the networking domain and used to analyze the energy-efficient path in the data or the node transfer from the base to the destination node. To find the best optimal energy-efficient path, they are classified using the DL techniques by implementing the modified Deep CNN-BiLSTM mechanism for organizing handover or the load balance technique for effective data transfer. The main reason for using DL techniques in this proposed method is to make the handoff of load-balancing techniques stand out and be organized for better node transfer mechanisms. This proposed method provides a low latency of 3.2s and an accuracy of 99.96% compared to existing methods.

Publisher

Springer Science and Business Media LLC

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3