Summary
International Symposium on Nonlinear Theory and its Applications
2008
Session Number:B4L-E
Session:
Number:B4L-E1
Autonomous and Decentralized Optimization for Fair Radio Resource Selection by Higher-Order Neural Networks
Taichi Takeda, Kuroda Taro, Mikio Hasegawa, Hiroshi Harada, Shuzo Kato, Yoshitoshi Murata,
pp.-
Publication Date:2008/9/7
Online ISSN:2188-5079
Summary:
Various advanced wireless systems have been developed and commercialized in recent years. In order to utilize them efficiently by switching among different wireless networks without interruption of an on-going session, vertical handover technologies have been developed and standardized. Moreover, cognitive radio networking technologies that optimize radio resource usage of a limited frequency band become also an important issue now. In order to optimize radio resource usage, we propose an autonomous and decentralized radio resource selection algorithm based on the optimization dynamics of the mutually-connected neural networks. The proposed neural network maximizes the average throughput per terminal and minimizes the differences of the throughput among the terminals at same time by using the fourth-order energy function. We show that the radio resource usage could be optimized by the proposed method based on decentralized and autonomous computing.