Summary
the 2014 International Symposium on Nonlinear Theory and its Applications
2014
Session Number:B1L-C
Session:
Number:B1L-C3
Efficient Routing Algorithm Using Mutually Connected Neural Networks withWaiting Transmitted Information
Tohru Takamizawa, Takayuki Kimura, Keisuke Kimura, Kenya Jin’no,
pp.245-248
Publication Date:2014/9/14
Online ISSN:2188-5079
Summary:
Because of the growth of the Internet users, data packets flowing in the computer networks become huge number, and the data packets are congested in the computer networks. If the packet congestion occurs in the computer networks, some packets are trapped into the congested nodes, then, these packets are delayed to be transmitted to the destinations. Further, the packets might be removed from the computer networks in the worst case. To overcome these undesirable problems, an efficient routing method which uses the mutually connected neural networks has been proposed by Horiguchi and Ishioka. This routing method shows good performance for the regular topology of the computer networks. However, the performance is degraded for the irregular topology of the networks. To improve the performance for the irregular networks, we propose a new routing method using mutually connected neural networks which considers the waiting transmitted information on the paths of the transmitted packets in this paper. From the results of the numerical experiments, the proposed method shows better performance than the conventional routing methods if the number of existing packets is small.