Application of fuzzy inference systems to detection of faults in wireless sensor networks

Khan, Safdar Abbas; Daachi, B.; Djouani, Karim

doi:10.1016/j.neucom.2012.04.002

Cited by 56 publications

(14 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[40] used principal component analysis with the neural network to determine the soft fault. [41] reported the fault detection using fuzzy inference system with its neighbor to detect the transient and permanent faults. These methods have drawbacks when the nodes are dynamic, resulting in inaccurate low level fault detection, and increased scalability of the network.…”

Section: Related Workmentioning

confidence: 99%

Hybrid Continuous Density Hmm-Based Ensemble Neural Networks for Sensor Fault Detection and Classification in Wireless Sensor Network

Emperuman

Chandrasekaran

2020

Sensors

View full text Add to dashboard Cite

Sensor devices in wireless sensor networks are vulnerable to faults during their operation in unmonitored and hazardous environments. Though various methods have been proposed by researchers to detect sensor faults, only very few research studies have reported on capturing the dynamics of the inherent states in sensor data during fault occurrence. The continuous density hidden Markov model (CDHMM) is proposed in this research to determine the dynamics of the state transitions due to fault occurrence, while neural networks are utilized to classify the faults based on the state transition probability density generated by the CDHMM. Therefore, this paper focuses on the fault detection and classification using the hybridization of CDHMM and various neural networks (NNs), namely the learning vector quantization, probabilistic neural network, adaptive probabilistic neural network, and radial basis function. The hybrid models of each NN are used for the classification of sensor faults, namely bias, drift, random, and spike. The proposed methods are evaluated using four performance metrics which includes detection accuracy, false positive rate, F1-score, and the Matthews correlation coefficient. The simulation results show that the learning vector quantization NN classifier outperforms the detection accuracy rate when compared to the other classifiers. In addition, an ensemble NN framework based on the hybrid CDHMM classifier is built with majority voting scheme for decision making and classification. The results of the hybrid CDHMM ensemble classifiers clearly indicates the efficacy of the proposed scheme in capturing the dynamics of change of statesm which is the vital aspect in determining rapidly-evolving instant faults that occur in wireless sensor networks.

show abstract

Section: Related Workmentioning

confidence: 99%

Hybrid Continuous Density Hmm-Based Ensemble Neural Networks for Sensor Fault Detection and Classification in Wireless Sensor Network

Emperuman

Chandrasekaran

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Fuzzy logic and fuzzy sets are powerful mathematical tools [33][34][35]. They are used for modeling and controlling uncertain systems in industry, humanity and nature.…”

Section: Fuzzy Logic Applicationsmentioning

confidence: 99%

“…Fuzzy decision making is only one of several useful decision making procedures which are based on mathematical proofs and presented in the literature [33][34][35]. However, the functional efficiency of the proposed protocol is also related to some parameters such as the number of nodes, initial energy of nodes, node buffer size and deploying method of the nodes in the whole network.…”

Section: Routing Procedures In the Proposed Protocolmentioning

confidence: 99%

Static Three-Dimensional Fuzzy Routing Based on the Receiving Probability in Wireless Sensor Networks

Gharajeh

Khanmohammadi

2013

Computers

View full text Add to dashboard Cite

A Wireless Sensor Network (WSN) is a collection of low-cost, low-power and large-scale wireless sensor nodes. Routing protocols are an important topic in WSN. Every sensor node should use a proper mechanism to transmit the generated packets to its destination, usually a base station. In previous works, routing protocols use the global information of the network that causes the redundant packets to be increased. Moreover, it leads to an increase in the network traffic, to a decrease in the delivery ratio of data packets, and to a reduction in network life. In this paper, we propose a new inferential routing protocol called SFRRP (Static Three-Dimensional Fuzzy Routing based on the Receiving Probability). The proposed protocol solves the above mentioned problems considerably. The data packets are transmitted by hop-to-hop delivery to the base station. It uses a fuzzy procedure to transmit the sensed data or the buffered data packets to one of the neighbors called selected node. In the proposed fuzzy system, the distance and number of neighbors are input variables, while the receiving probability is the output variable. SFRRP just uses the local neighborhood information to forward the packets and is not needed by any redundant packet for route discovery. The proposed protocol has some advantages such as a high delivery ratio, less delay time, high network life, and less network traffic. The performance of the proposed protocol surpasses the performance of the Flooding routing protocol in terms of delivery ratio, delay time and network lifetime

show abstract

“…Then the service properties like network coverage and connectivity are greatly decreased. So a fault-tolerant topology is needed to continuously provide critical network services in the condition of nodes failure [1]. In order to evaluate fault tolerance of this topology, it is significant to establish the fault tolerance evaluation index of topology.…”

Section: Introductionmentioning

confidence: 99%

A quantitative fault tolerance evaluation model for topology in wireless sensor networks based on the semi-Markov process

Yin

Liu

et al. 2015

Neurocomputing

View full text Add to dashboard Cite

Application of fuzzy inference systems to detection of faults in wireless sensor networks

Cited by 56 publications

References 21 publications

Hybrid Continuous Density Hmm-Based Ensemble Neural Networks for Sensor Fault Detection and Classification in Wireless Sensor Network

Hybrid Continuous Density Hmm-Based Ensemble Neural Networks for Sensor Fault Detection and Classification in Wireless Sensor Network

Static Three-Dimensional Fuzzy Routing Based on the Receiving Probability in Wireless Sensor Networks

A quantitative fault tolerance evaluation model for topology in wireless sensor networks based on the semi-Markov process

Contact Info

Product

Resources

About