This paper is devoted to the dependability assessment of industrial wireless communication. The objective of this paper is to find the most significant assessment parameters and functions of logical link. First, motivation is given for new approaches to the dependability assessment of industrial networks. Next, the approach of the up-state function (USF) is introduced. For the construction of USF, significant parameters are needed, which are sought by different methods such as principal component analysis (PCA) and regression analysis. Finding the optimal variables for every environment and condition helps to identify the factors affecting the quality of wireless communication. In this paper, we are employing time series analysis and recurrent neural network to predict the most reliable value for each parameter for the next state. The analysis is based on a numerical training data set made of five different correlated parameters, including transmission time, update time, consecutive message loss, sent messages and lost messages.
[1]
Lutz Rauchhaupt,et al.
Methodology for holistic assessment of dependability in wireless automation
,
2017,
2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETFA).
[2]
Herbert Jaeger,et al.
Reservoir computing approaches to recurrent neural network training
,
2009,
Comput. Sci. Rev..
[3]
Richard A. Davis,et al.
Introduction to time series and forecasting
,
1998
.
[4]
Yunpeng Wang,et al.
Long short-term memory neural network for traffic speed prediction using remote microwave sensor data
,
2015
.
[5]
Prajakta S. Kalekar.
Time series Forecasting using Holt-Winters Exponential Smoothing
,
2004
.
[6]
Nasser M. Nasrabadi,et al.
Pattern Recognition and Machine Learning
,
2006,
Technometrics.
[7]
Robert Fildes,et al.
Principles of Business Forecasting
,
2012
.