Detection In Wireless Sensor Networks Research Articles

Standalone noise and anomaly detection in wireless sensor networks: A novel time‐series and adaptive Bayesian‐network‐based approach

SummaryWireless sensor networks (WSNs) consist of small sensors with limited computational and communication capabilities. Reading data in WSN is not always reliable due to open environmental factors such as noise, weakly received signal strength, and intrusion attacks. The process of detecting highly noisy data is called anomaly or outlier detection. The challenging aspect of noise detection in WSN is related to the limited computational and communication capabilities of sensors. The purpose of this research is to design a local time‐series‐based data noise and anomaly detection approach for WSN. The proposed local outlier detection algorithm (LODA) is a decentralized noise detection algorithm that runs on each sensor node individually with three important features: reduction mechanism that eliminates the noneffective features, determination of the memory size of data histogram to accomplish the effective available memory, and classification for predicting noisy data. An adaptive Bayesian network is used as the classification algorithm for prediction and identification of outliers in each sensor node locally. Results of our approach are compared with four well‐known algorithms using benchmark real‐life datasets, which demonstrate that LODA can achieve higher (up to 89%) accuracy in the prediction of outliers in real sensory data.

Joint PCA and Adaptive Threshold for Fault Detection in Wireless Sensor Networks

Joint PCA and Adaptive Threshold for Fault Detection in Wireless Sensor Networks

Minimum Byzantine Effort for Blinding Distributed Detection in Wireless Sensor Networks

In this work, we consider the general problem of distributed detection in the presence of Byzantines using wireless sensor networks. Instead of attempting to mitigate Byzantine attacks as a system designer, we investigate the issue from the perspective of a Byzantine attacker. The probability for each individual sensor to be compromised (compromising probability) required to blind the system operation is adopted as the attack measure. Under the system setting that the fusion center (FC) declares the most likely hypothesis to be true based on the $M$ -ary data from $N$ local sensors, a Byzantine attack policy that can blind the FC with the minimum compromising probability for each individual sensor is derived under the assumption that the Byzantine attacker knows the statistics of the local outputs. The closed-form expression for a blind-achieving Byzantine transition probability that is used to alter the statistics of the local outputs of compromised sensors is also established. Our results indicate that the statistics of the local outputs is essential for the minimization of an attacker's effort.

A mobile fault detection algorithm in heterogeneous wireless sensor networks: a bio-inspired approach

This paper puts forth a novel mobile fault detection algorithm for wireless sensor networks (WSNs) based on bacterial-inspired optimization. We introduce a bio-swarm intelligence approach to mobile fault detection in WSNs by using voltage values. At certain times, the sensor nodes in the clustered network send data packets containing health-fitness information to cluster heads (CHs) selected by the proposed CH selection algorithm. A mobile sink (MS) collects the health status via data from all the nodes as they reach the intersection point of the CHs. After this stage, the data packets are analyzed by the MS, and hardware or software faults are detected by assessing the fitness values of the nodes. The faulty nodes are eventually discarded from the network, and recovery of the rest of the nodes in the network is satisfied. Inspired by the interaction of bacteria for feed collection, their response to chemicals, and their interaction and communication with one another, we bring an innovative approach to finding node failures or software faults in WSNs, and these failures are removed from the network to help its operation and to take measures to maintain the electrical structures. In fact, we adapt our algorithm to low energy harvesting electrical components as an example. We compare our novel algorithm with existing studies through extensive simulations in NS 2 environment based on fault detection accuracy, false alarm rate, and false positive rate criteria versus fault probability, number of nodes, and sink speed. Considering detection accuracy, the simulation results validate that our algorithm shows better performance as compared with others.

Modelling of wireless sensor networks for detection land and forest fire hotspot

Indonesia located in South East Asia countries with tropical region, forest fires in Indonesia is one of big issue and disaster because it happens in almost of every year, this is because of some of region consist of peat land that high risk for fire especially in dry season. Riau Province is one of region that regularly incident of forest fire with affected the length and breadth of Indonesia. Propose development of Wireless Sensor Networks (WSNs) for detection of land and forest fire hotspot in Indonesia as well as one of the main consents in this research, case location in Riau province is at one of the regions that high risk forest fire in dry season. WSNs technology used for ground sensor system to collect environmental data. Data training for fire hotspot detection is done in data center to determine and conclude of fire hotspot then potential to become big fire. The deployment of sensors located at several locations that has potential for fire incident, especially as data shown in previous case and forecast location with potential fire happen. Mathematical analysis is used in this case for modelling number of sensors required to deploy and the size of forest area. The design and development of WSNs give high impact and feasibility to overcome current issues of forest fire and fire hotspot detection in Indonesia. The development of this system used WSNs highly applicable for early warning and alert system for fire hotspot detection.

An Adaptive Framework for Event Detection in Wireless Sensor Network

A wireless sensor network is the new generation technology that holds the capability and it has transformed to interact with the physical environment according to the user’s demands. The features of a sensor network play the major role in various application fields that fit in the broad category for tracking, monitoring, automation and detection and in that category WSN plays the predominant role in event monitoring. Event detection in sensor networks is the process of observing; evaluating an event based on multiple attributes and generates an alarm at the appropriate time. A framework was developed to detect the event in in-door environment. The proposed event detection model comprises of three phases. In the first phase the network id grouped and cluster head is determined and subsequently composite model was developed to determine the event and the final phase involves in the fuzzy rule optimization. A simulation setup has created and the experimental evaluation validates the event detection mechanism through various metrics such as cluster communication, event accuracy are measured and evaluated. The results are compared with the energy efficient HEED algorithm and connected dominating set based cluster MCDS-CCH algorithm further the composite event detection mechanism with optimization rule outperforms than the well-established J48 decision tree classification algorithm.

Behavior based Routing Misbehavior Detection in Wireless Sensor Networks

Behavior based Routing Misbehavior Detection in Wireless Sensor Networks

Globalized robust Markov perfect equilibrium for discounted stochastic games and its application on intrusion detection in wireless sensor networks: Part I—theory

In this article, we study a discounted stochastic game to model resource optimal intrusion detection in wireless sensor networks. To address the problem of uncertainties in various network parameters, we propose a globalized robust game-theoretic framework for discounted robust stochastic games. A robust solution to the considered problem is an optimal point that is feasible for all realizations of data from a given uncertainty set. To allow a controlled violation of the constraints when the parameters move out of the uncertainty set, the concept of globalized robust framework comes into view. In this article, we formulate a globalized robust counterpart for the discounted stochastic game under consideration. With the help of globalized robust optimization, a concept of globalized robust Markov perfect equilibrium is introduced. The existence of such an equilibrium is shown for a discounted stochastic game when the number of actions of the players is finite. The contraction mapping theorem, Kakutani fixed point theorem and the concept of equicontinuity are used to prove the existence result. To compute a globalized robust Markov perfect equilibrium for the considered discounted stochastic game, a tractable representation of the proposed globalized robust counterpart is also provided. Using the derived tractable representation, we formulate a globalized robust intrusion detection system for wireless sensor networks.

New trends in adaptive smart systems, services, and architectures

Smart Systems are composed of interconnected nodes (devices, services, components, etc.) that can be intelligent, mobile, embedded, cloud-based, or/and adaptive. These nodes interact together through new communication protocols. Architectures of such system needs to adapt its behavior and deals with change in the environment. An adaptive and reconfigurable architecture can repair itself if any execution problems occur, in order to successfully complete its own execution. In the design of an adaptive and reconfigurable software system, several aspects have to be considered. Also other emerging areas such as Big data can also be used to solve traditional problems of Smart Systems, like scalability and performance. This special issue addresses these issues and is composed of seven papers. The paper of1 entitled “Efficient monitoring for intrusion detection in wireless sensor networks” addresses the design of a reliable and energy-efficient monitoring system that resists internal security attacks based on a distributed smart monitoring protocol. The paper of2 entitled “Cloud SLA negotiation and re-negotiation: An ontology-based context-aware approach” defines a context-aware system that automates the negotiation of dynamically generated and adapted SLAs using reasoning techniques and semantic mapping between client requests and provider offers. The paper of3 entitled “Cloud service composition using minimal unsatisfiability and genetic algorithm” targets the run-time phase of adaptive systems. It addresses the dynamic properties of service-oriented adaptive systems by provide a hybrid approach for automatic WS composition based on formal methods and genetic algorithms. The paper of4 entitled “A knowledge-based approach to manage configurable business processes” targets both run-time and design-time as well as structural and behavioral properties of enterprise systems modelled and enacted as Business Processes (BP). It proposes a knowledge-based approach for capturing the different versions of a BP into a single configurable BP and splitting this process into fragments as an alternative of handling multiple BP versions. The paper of5 entitled “Avoiding resource misallocations in business processes” presents an approach for avoiding resource misallocations, considered as an improper combination of resources, in Business Processes (BPs). The paper of6 entitled “Simulating systems of systems using situation/reaction paradigm” targets the designs-time phase of adaptive systems. It considers the paradigm of Systems of Systems that can be applied to different smart and adaptive systems. The paper sees these systems from the mission point of view to characterise their evolving rules with formal modelling and simulation. The paper of7 entitled “A comprehensive survey on modeling of cyber-physical systems” provides a survey of existing approaches to modeling CPS. It aims to study and to classify the CPS properties and to discuss their importance in different application domains. We would like to thank CCPE Editor-in-Chief Dr. Geoffrey C. Fox for his strong support. A special thank goes to all authors for their valuable contributions to this special issue. Special thanks go to all the guest eidtorial board members for their comments that helped in enhancing the quality of the papers.

Anomaly detection in wireless sensor networks based on time factor

Anomaly detection in wireless sensor networks based on time factor

Congestion avoidance and fault detection in WSNs using data science techniques

AbstractTransmission rate is one of the contributing factors in the performance of wireless sensor networks. Congested network causes reduced network response time, queuing delay, and more packet loss. To address the issue of congestion, we have proposed transmission rate control methods. To avoid the congestion, we have adjusted the transmission rate at current node based on its traffic loading information. Multiclassification is done to control the congestion using an effective data science technique, namely support vector machine (SVM). In order to get less miss classification error, differential evolution (DE) and grey wolf optimization (GWO) algorithms are used to tune the SVM parameters. The comparative analysis has shown that the proposed approaches DE‐SVM and GWO‐SVM are more proficient than other classification techniques. Moreover, DE‐SVM and GWO‐SVM have outperformed the benchmark technique genetic algorithm–SVM by producing 3% and 1% less classification errors, respectively. For fault detection in wireless sensor networks, we have induced four types of faults in the sensor readings and detected the faults using the proposed enhanced random forest. We have made a comparative analysis with state of the art data science techniques based on two metrics, ie, detection accuracy and true positive rate. Enhanced random forest has detected the faults with 81% percent accuracy and outperformed the other classifiers in fault detection.

A Genetic-Based Extreme Gradient Boosting Model for Detecting Intrusions in Wireless Sensor Networks

An Intrusion detection system is an essential security tool for protecting services and infrastructures of wireless sensor networks from unseen and unpredictable attacks. Few works of machine learning have been proposed for intrusion detection in wireless sensor networks and that have achieved reasonable results. However, these works still need to be more accurate and efficient against imbalanced data problems in network traffic. In this paper, we proposed a new model to detect intrusion attacks based on a genetic algorithm and an extreme gradient boosting (XGBoot) classifier, called GXGBoost model. The latter is a gradient boosting model designed for improving the performance of traditional models to detect minority classes of attacks in the highly imbalanced data traffic of wireless sensor networks. A set of experiments were conducted on wireless sensor network-detection system (WSN-DS) dataset using holdout and 10 fold cross validation techniques. The results of 10 fold cross validation tests revealed that the proposed approach outperformed the state-of-the-art approaches and other ensemble learning classifiers with high detection rates of 98.2%, 92.9%, 98.9%, and 99.5% for flooding, scheduling, grayhole, and blackhole attacks, respectively, in addition to 99.9% for normal traffic.

Exploiting Spatial Proximity for Detection in Wireless Sensor Network

In a large-scale wireless sensor network, weak event signals spread over a certain region that contains a subset of sensors located in close proximity, thus forming a spatial cluster around the event. Only this subset of sensors receives credible information from the event and information of the remaining far away sensors will be dominated by background noise, therefore less credible and faulty. We exploit this fact for improving the performance of multi-sensor fusion rule by proposing novel techniques that give importance to the nearby sensors and utilize their proximity information. The proposed mechanism is demonstrated to improve the performance of fusion rule when (1) fusion center receives soft information from the local sensors and (2) when sensors send binary hard information in the form of 1/0 to the fusion center. For the assessment of the proposed schemes, simulation results are provided to show that the proposed schemes have better performance compared to the traditional schemes.

Secured Data Transmission and Malicious Node Detection in Wireless Sensor Network

Wireless Sensor Network (WSN) is developed extremely because of their low installation cost and various applications. WSN has compact and inexpensive sensor nodes for monitoring the physical environment. WSNs are susceptible to many attacks (e.g. malicious nodes) because of its distinct characteristics. The performance of node and network is affected by the malicious nodes. Moreover, the communication among the sensor nodes also required to be secured for preventing the data from the hackers. In this paper, the architecture of the WSN is generated by using the Fuzzy-C-Means clustering (FCM). Then the detection of the malicious nodes is performed by using the Acknowledgement Scheme (AS). This AS is integrated in the Ant Colony Optimization (ACO) based routing for avoiding the malicious nodes while generating the route from the source to the Base Station (BS). Then the Hybrid Encryption Algorithm (HEA) is used for performing the secure data transmission through the network and this proposed method is named as HEA-AS. The performance of the HEA-AS method is evaluated in terms of End to End Delay (EED), network lifetime, throughput, Packet Delivery Ratio (PDR) and Packet Loss Ratio (PLR). The proposed HEA-AS method is compared with the existing method called as CTCM to evaluate the effectiveness of the HEA-AS method.

Signal Detection from Permutated Observations Using Distributed Sensors

In this paper, distributed constant level detection in wireless sensor networks (WSNs) is investigated. The permuted linear model with a scalar parameter and additive heteroscedastic Gaussian noise is introduced, where the associations between the observations and the sensors are assumed to be unknown. Several detectors such as the approximations of the generalized likelihood ratio test (GLRT) detector, mean detector, and energy detector are proposed, and their receiver operating characteristics (ROCs) are evaluated. Numerical simulations are performed, and it is shown that the performance degradation of the GLRT detector is small, compared to the permutation known as Neyman-Pearson (NP) detector.

Trust evaluation model with entropy-based weight assignment for malicious node\u2019s detection in wireless sensor networks

Trust management is considered as an effective complementary mechanism to ensure the security of sensor networks. Based on historical behavior, the trust value can be evaluated and applied to estimate the reliability of the node. For the analysis of the possible attack behavior of malicious nodes, we proposed a trust evaluation model with entropy-based weight assignment for malicious node’s detection in wireless sensor networks. To mitigate the malicious attacks such as packet dropping or packet modifications, multidimensional trust indicators are derived from communication between adjacent sensor nodes, and direct and indirect trust values will be estimated based on the corresponding behaviors of those sensor nodes. In order to improve the validity of trust quantification and ensure the objectivity of evaluation, the entropy weight method is applied to determine the proper value of the weight. Finally, the indirect trust value and direct trust value are synthesized to obtain the overall trust. Experimental results show that the proposed scheme performs well in terms of the identification of malicious node.

Data Mining Based Intrusion Detection in Wireless Sensor Network

Intrusion detection is the one of the challenging task in wireless sensor network and prevents the system and network resources from being intrude or compromised. One of the ongoing strategies for recognizing any anomalous activities presented in a network is done by intrusion detection systems (IDS) and it becomes an essential part of defense system against attacker problems. The primary goal of our work is to study and analyze intrusion detection technique meant for improving the performance of Intrusion Detection using hybrid ANN based Clustering technique. To estimate the effectiveness of the proposed strategy, KDD CUP 99 dataset is utilized for testing and assessment. Based on the analysis, it is noticed that the proposed ANN clustering performs much better than other methods with respect to accuracy which attains an average high accuracy of 93.91%when compared with other methods.

Artificial bee colony based sinkhole detection in wireless sensor networks

Sinkhole attack in wireless sensor networks (WSN) is most vulnerable attack in WSN that prevents the base station from gathering complete and unmodified data from its origin. A simple authentication mechanism is not adequate to prevent WSN from sinkhole attacks as signed routing can also be easily done by compromised nodes. Hence in this paper we addressed the problem sinkhole attack detection in WSN using swarm-based algorithm namely artificial bee colony algorithm. This algorithm finds the compromised node by comparing the node ID’s defined in the rule set. ABC reduces the overall time complexity taken to find the compromised node which turns to reduces the packet loss percentile and increases the packet delivery ratio. The performance of the proposed algorithm is evaluated and compared with the existing methodologies. The results show that the proposed algorithm outperforms the existing methodologies in terms of packet loss, packet delivery ratio and energy consumption.

Temporal and spatial outlier detection in wireless sensor networks

Outlier detection techniques play an important role in enhancing the reliability of data communication in wireless sensor networks (WSNs). Considering the importance of outlier detection in WSNs, many outlier detection techniques have been proposed. Unfortunately, most of these techniques still have some potential limitations, that is, (a) high rate of false positives, (b) high time complexity, and (c) failure to detect outliers online. Moreover, these approaches mainly focus on either temporal outliers or spatial outliers. Therefore, this paper aims to introduce novel algorithms that successfully detect both temporal outliers and spatial outliers. Our contributions are twofold: (i) modifying the Hampel Identifier (HI) algorithm to achieve high accuracy identification rate in temporal outlier detection, (ii) combining the Gaussian process (GP) model and graph‐based outlier detection technique to improve the performance of the algorithm in spatial outlier detection. The results demonstrate that our techniques outperform the state‐of‐the‐art methods in terms of accuracy and work well with various data types.

On semantic clustering and adaptive robust regression based energy-aware communication with true outliers detection in WSN

To conserve energy and enhance the lifetime of the wireless sensor network (WSN), reducing the amount of data communication by exploiting temporal and spatial correlation of sensed data is well suitable technique. So, instead of sending every data to the destination, it can be worthy of introducing a prediction method to reduce redundant data transmission by exploiting the temporal correlation of sensed data. We show that the prediction accuracy of source data depends not only on the method applied but also on the correctness of the sample data provided by the source nodes. Erroneous sample data (outliers) leads to the wrong prediction. In this paper, we propose an energy efficient SEMantic CLustering (SEMCL) model to mitigate high energy consumption problem in a clustered WSN. Our model produces energy efficient clusters by strong intra-cluster data similarity to exploit spatial correlation of data. We adopt the Robust and Efficient Weighted Least Square method (REWLS) to provide accurate data prediction with negligible errors. Because REWLS method lacks to differentiate true and false outliers and thus to improve further the Quality of Service (QoS) on data accuracy, we propose a separate algorithm, named, True Outlier Detection (TOD). Moreover, to improve the QoS in communications, a reliable backbone network based on the link quality of the data forwarding path has been implemented. Our proposed model has been simulated using real data and compared with the existing techniques to show its efficacy and superiority in terms of QoS on data accuracy, energy consumption, and network lifetime.