Congestion Avoidance on Networks Using Independent Memory Information

Kimura, Takayuki

doi:10.1007/978-3-030-10892-2_17

Cited by 3 publications

(10 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[26] by improving the number of transmissions of packets and buffer sizes at each node. To show that the conventional communication network models [23,24,26] have poor packet routing abilities, we first displayed relationships between degree and node betweenness centralities of networks and the number of existing packets by using the shortest path (SP) method. The SP method transmits packets from sources to destinations using uniquely determined shortest paths.…”

Section: Communication Network Modelsmentioning

confidence: 99%

“…In all the numerical experiments, we set the number of iterations, denoted by I, as I = 10 3 . In addition, we set β and γ to 10 3 and 0.4, respectively, based on conventional studies [23,24]. We also generated R packets at each iteration.…”

Section: Communication Network Modelsmentioning

confidence: 99%

“…In the routing method using the chaotic neurodynamics, the refractory effect works to memorize the histories of packet transmissions at each node. Further, a routing method using memory information, which reinforces the packet memorizing functionality in chaotic neural networks, has already been proposed [23]. According to the hop distance information and memory information, this routing method [23] avoids the congestion of packets and maintains a high transmission completion rate of packets for the communication networks with SF properties.…”

Section: Introductionmentioning

confidence: 99%

“…Further, a routing method using memory information, which reinforces the packet memorizing functionality in chaotic neural networks, has already been proposed [23]. According to the hop distance information and memory information, this routing method [23] avoids the congestion of packets and maintains a high transmission completion rate of packets for the communication networks with SF properties. In addition, the effectiveness of the memory information is clarified by using the method of surrogate data [24], which is a statistical hypothesis testing used in the research field of nonlinear time-series analysis [25].…”

Section: Introductionmentioning

confidence: 99%

“…In this work, we incorporated two strategies to improve the routing method using memory information [23,24]. First, we reformed the conventional communication network models [26] by changing the number of transmissions of packets and buffer sizes at each node based on node betweenness centrality, which is one of the network centralities that measures the number of shortest paths that pass through nodes in the networks [27].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A routing method with adaptively adjusting memory information based on local routing history

Kimura

Shimada

2023

PLoS ONE

Self Cite

View full text Add to dashboard Cite

Finding the shortest paths for packets from sources to destinations in packet-switched communication networks is an inevitable problem in building a future high-speed information society. A routing method with memory information has already been proposed to alleviate the congestion of large volumes of packet flow. This routing method shows a high transmission completion rate even for large volumes of packet flows in communication networks with scale-free properties. However, the method exhibits poor performance for networks with local triangular connections and long distances between nodes. To overcome these problems, in this study, we first enhanced the routing performance of the conventional communication network models by using the betweenness centrality of nodes, which is one of the network centralities that measures the number of shortest paths that pass through each node in the networks. Subsequently, we adaptively changed the transmitting paths of packets by using only local information. Numerical simulations indicated that our routing method performs successfully for various topologies of communication networks by avoiding the congested node, and effectively using the memory information.

show abstract