A CNN-based network failure prediction method with logs

Ji, Weiliang; Duan, Shihui; Chen, Renai; Wang, Song; Ling, Qiang

doi:10.1109/ccdc.2018.8407833

Cited by 11 publications

(3 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They can be used for short-term prediction of network failures. In the case of the methods described in [31,48], their proper functioning requires the maximally extended logging of traffic and activity of network nodes. This approach in the MESH networks considered in the article due to the assumed battery power supply would significantly reduce the network lifetime and significantly reduce the value of the available transmission band for the relevant data (other than diagnostic and control packages).…”

mentioning

confidence: 99%

Phase Transitions in Wireless MESH Networks and Their Application in Early Detection of Network Coherence Loss

Paszkiewicz¹,

Bolanowski²,

Zapała³

2019

Applied Sciences

View full text Add to dashboard Cite

The paper focuses on the phenomenon of phase transitions related to changes taking place directly in a network structure; i.e., in physical topology. Although these changes are related to the physical layer, they directly affect the functioning of higher layers, and thus the quality and efficiency of network transmission. The research presented herein was focused in particular on phase transformations in MESH type wireless topologies. The dependence between the values of radius (transmitter power) nodes and the coherence of the MESH network is presented. These considerations are important for understanding the dynamics of processes occurring in topological structures, and for optimizing the use of power sources for individual nodes. The paper also presents the use of phase transition analysis for early detection of network coherence loss at the physical layer level. A new mechanism for assessing the level of network communication and an active method of preventing its failure using the arbiter and a set of limited diagnostic information are described. The solution presented can be used both in Internet of Things (IoT) systems and Industry 4.0 to create a high level of reliability wireless connection infrastructure between mobile and fixed devices.

show abstract

mentioning

confidence: 99%

Phase Transitions in Wireless MESH Networks and Their Application in Early Detection of Network Coherence Loss

Paszkiewicz¹,

Bolanowski²,

Zapała³

2019

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…Convolution simply does not properly represent the irregular structure of a network. The second case of stepping through log files (Ji et al, 2018) is less clear in the assumption of structure, but the third case (Xiao et al, 2019), where feature vectors are reshaped into matrices, places the features in arbitrary and potentially uncorrelated relation to one another.…”

Section: Machine Learning On Networkmentioning

confidence: 99%

“…MLPs have also been used (Srinivasan, Truong-Huu, and Gurusamy, 2019;Rafique et al, 2018;Feng et al, 2018). There are also CNN-based approaches (Wang et al, 2018b;Ji et al, 2018;Xiao et al, 2019), some of which, as we have seen in Section 2.6.1, force data into grid-like structures and therefore are questionable with respect to the structural assumptions implicit in a CNN (Wang et al, 2018b;Xiao et al, 2019).…”

Section: Introductionmentioning

confidence: 99%