Symbolic time series analysis for anomaly detection: A comparative evaluation

Chin, Shin; Ray, Asok; Rajagopalan, Venkatesh

doi:10.1016/j.sigpro.2005.03.014

Cited by 63 publications

(36 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A comparative evaluation of this novel analytical method has shown its superior performance relative to other existing pattern recognition tools in terms of early detection of anomalies [20][21][22] and robustness to noisy environments [23]. While the details are reported in previous publications [16,23,24], a brief review of the essential concepts of space partitioning and symbol sequence generation is presented here for self-sufficiency, clarity and completeness of the paper.…”

Section: A Brief Review Of Sdfmentioning

confidence: 99%

See 1 more Smart Citation

Statistical pattern analysis of ultrasonic signals for fatigue damage detection in mechanical structures

Gupta

Singh

Ray

2008

NDT & E International

Self Cite

View full text Add to dashboard Cite

Section: A Brief Review Of Sdfmentioning

confidence: 99%

“…(c) Capability for early detection of anomalies because of enhanced sensitivity to signal distortion using escort distributions. (d) Real-time execution on commercially available inexpensive platforms [20,21].…”

Section: Escort Distributions For Pattern Identificationmentioning

confidence: 99%

Statistical pattern analysis of ultrasonic signals for fatigue damage detection in mechanical structures

Gupta

Singh

Ray

2008

NDT & E International

Self Cite

View full text Add to dashboard Cite

“…D-Markov model is presented by Ray to detect anomaly behaviour in mechanical and electronic system. In recent years, Ray and his group discuss the essence of D-Markov model with symbolic dynamics (Ray 2004), evaluate its performance comparing with existing techniques (Chin, Ray, and Rajagopalan 2005), develop a unique symbolisation method, wavelet-based partitioning, to improve the capability of the model (Rajagopalan and Ray 2006) and identify the critical system parameters as an inverse problem (Rajagopalan, Cakraborty, and Ray 2008). The effect of the anomaly detection algorithm Given a symbol series Á Á Á S À2 S À1 S 0 S 1 S 2 Á Á Á, its D-Markov model can be constructed as following.…”

Section: Epsilon Machine and D-markov Machinementioning

confidence: 99%

Anomaly detection in complex system based on epsilon machine

Xiang¹,

Zhou²,

Jiang³

2008

International Journal of Systems Science

View full text Add to dashboard Cite

Epsilon machine is a computational mechanics theory and its most effective reconstruction algorithm is causal state splitting reconstruction (CSSR). As CSSR can only be applied to symbol series, symbolising real series to symbol series is necessary in practice. Epsilon machine discovers the hidden pattern of a system. In reconstructed results, the hidden pattern is expressed as the set of causal states. Based on the variation of causal states, a novel anomaly detection algorithm, structure vector model, is presented. The vector is composed of the causal states, and the anomaly measure is defined with the distance of different vectors. An example of the crankshaft fatigue demonstrates the effectiveness of the model. The mechanism of the model is discussed in detail from three aspects, computational mechanics, symbolic dynamics and complex networks. The new idea defining anomaly measure based on the variation of hidden patterns can be interpreted reasonably with the hierarchical structure of complex networks. The jump in anomaly curves is a nature candidate for the threshold, which confirms the positive meaning of the model. Finally, the parameter choice and time complexity are briefly analysed.

show abstract

“…This section presents the experimental results to validate the concept of STSA-based pattern identification on a laboratory apparatus with computer instrumented electronic systems [5,6]. Two nonlinear systems described by Duffing equation and Van der Pol equation [11] are considered.…”

Section: Experimental Validationmentioning

confidence: 99%

“…This methodology was compared with other data driven methods such as principal component analysis (PCA) and artificial neural networks (ANN) in Ref. [6]. A new partitioning scheme to enhance this methodology, known as maximum entropy (ME) partitioning, was reported in Ref.…”

mentioning

confidence: 99%

Pattern identification in dynamical systems via symbolic time series analysis

Rajagopalan¹,

Ray²,

Samsi³

et al. 2007

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

This paper presents symbolic time series analysis (STSA) of multi-dimensional measurement data for pattern identification in dynamical systems. The proposed methodology is built upon concepts derived from Information Theory and Automata Theory. The objective is not merely to classify the time series patterns but also to identify the variations therein. To achieve this goal, a symbol alphabet is constructed from raw data through partitioning of the data space. The maximum entropy method of partitioning is extended to multi-dimensional space. The resulting symbol sequences, generated from time series data, are used to model the dynamical information as finite state automata and the patterns are represented by the stationary state probability distributions. A novel procedure for determining the structure of the finite state automata, based on entropy rate, is introduced. The diversity among the observed patterns is quantified by a suitable measure. The efficacy of the STSA technique for pattern identification is demonstrated via laboratory experimentation on nonlinear systems.

show abstract

Symbolic time series analysis for anomaly detection: A comparative evaluation

Cited by 63 publications

References 3 publications

Statistical pattern analysis of ultrasonic signals for fatigue damage detection in mechanical structures

Statistical pattern analysis of ultrasonic signals for fatigue damage detection in mechanical structures

Anomaly detection in complex system based on epsilon machine

Pattern identification in dynamical systems via symbolic time series analysis

Contact Info

Product

Resources

About