权益分类	功能权益	普通用户	{{item.name}}会员
{{category.name}}	{{benefitItem.name}}

Research on Neural Network Based Dynamic Channel Assignment Systems in Cellular Mobfle Communication Systems

蜂窝移动通信系统中基于神经网络的动态信道分配系统的研究

基本信息

批准号：
10555126
负责人：
SENGOKU Masakazu
金额：
$ 7.94万
依托单位：
NIIGATA UNIVERSITY
依托单位国家：
日本
项目类别：
Grant-in-Aid for Scientific Research (B)
财政年份：
1998
资助国家：
日本
起止时间：
1998 至 2001
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-10555126/
关键词：
Mobile communication Cellular system Channel assignment Base station location Cell location Neural net Genetic algorithm Optimization methods チャネル割当て最適化問題

项目摘要

Because demand for mobile communication services rapidly increases, the efficient use of frequency bands is an important issue. For the efficient use of the frequency bands, mobile communication networks should be highly optimized. Also, it is neciessary to apply intelligent control methods to mobile communication networks. To optimize the mobile networks, we must solve some NP complete problems quiclfly. For this purpose,neural networks and generic algorithms can be used, and it is very important to develop quick dynamic channel assignment systems using neural networks and genetic algorithms. The objective of this project is to develop dynamic channel assignment systems using neural networks and genetic algorithms, and to evaluate performance of the systems. It is considered that we can apply parallel computation systems consisting of some CPUs to dynamic channel assignment systems. Hence, this project develops dynamic channel assignment systems based on neural networks or genetic algorithms with parallel computation. This project has developed such systems for dynamic channel assignment. Through the evaluation process, we confirmed performance improvement by using the developed method. Also, we examined channel assignment methods to adapt to traffic fluctuation caused by vehicles' traffic fluctuation. Also, this project developed an approximate method to theoretically analyze communication traffic characteristics of dynamic channel assignment methods considering mobility. Other optimization problems and fundamental problems to improve mobile communication systems were also considered.

由于对移动的通信服务的需求迅速增加，频带的有效使用是一个重要问题。为了有效地使用频带，移动的通信网络应该被高度优化。此外，将智能控制方法应用于移动的通信网络也是必要的。为了优化移动的网络，必须快速地解决一些NP完全问题.为此，可以使用神经网络和遗传算法，并且使用神经网络和遗传算法开发快速动态信道分配系统是非常重要的。本计画的目标是利用类神经网路与基因演算法来发展动态频道分配系统，并评估系统的效能。本文认为可以将由多个CPU组成的并行计算系统应用于动态信道分配系统。因此，本计画将发展以平行运算之神经网路或基因演算法为基础之动态频道分配系统。该项目已开发了动态信道分配系统。通过评估过程中，我们证实了使用开发的方法的性能改进。此外，我们研究了信道分配方法，以适应交通波动引起的车辆的交通波动。此外，本项目开发了一种近似方法，从理论上分析考虑移动性的动态信道分配方法的通信业务特性。还考虑了改进移动的通信系统的其它优化问题和基本问题。