FAULT DETECTION IN WIRELESS SENSOR NETWORKS

Transcription

1 FAULT DETECTION IN WIRELESS SENSOR NETWORKS Manisha 1, Mr. Deepak Nandal 2 1 M.Tech (CSE), GJUS&T, Hisar, India 2 Assistant Professor, GJUS&T, Hisar, India ABSTRACT Wireless Sensor Networks are infrastructures containing sensing, computing and communication elements that aim to give its controllers the ability to measure, collect and react to occurrences in the monitored environment.wsn focus on interaction with environment. Nothing is perfect in this universe. WSN may also be failure prone. In order to maintain the network quality of service, it is mandatory for WSN to be able to detect the faults and take appropriate actions to handle them. The main aim of this article is to give an introduction with faults in WSN, and their detection approaches. Also it provide a place where one can have interaction with some efficient fault detection algorithms such as DFD scheme, localized fault detection algorithm and CDFD algorithm etc. Keywords: Fault Detection, WSN 1. INTRODUCTION Development of WSN is truly fitted in the famous proverb Necessity is the Mother of Invention. As the technology is getting very advanced, especially in the field of Electro-Micro-Mechanical systems has expedited the development of smart sensors. Due to this, it is possible to create a network by connecting independent sensor nodes. Wireless Sensor Network is applicable in various fields such as data acquisition in hazardous environment, monitoring of critical infrastructures and military operations. The unfriendly environment affects the monitoring infrastructure of wireless sensor networks (WSNs). Sensor nodes are expected to operate autonomously in unattended and possibly hostile environments. The lifetime of sensor node may vary from few hours to months or years depends upon the context in which it runs. Thus WSNs are vulnerable to faults where faults are likely to occur frequently and unexpectedly. As faults are unavoidable in the sensor network, it is very necessary to distinguish between faulty and working nodes. For all of these reasons, fault management is a major design challenge in WSN. Faults must be handled with extra attention and care. 2. RELATED WORK 2.1 How It Works? A WSN can be defined as a network of devices, denoted as nodes, which can sense the environment and communicate the information gathered from the monitored field through wireless links. The data is forwarded, possibly via multiple hops, to a sink (sometimes denoted as controller or monitor) that can use it locally or is connected to other networks (e.g., the Internet) through a gateway. The nodes can be stationary or moving [1]. WSNs are typically selforganizing and self-healing. Self-organizing networks allow a new node to automatically join the network without the need for manual intervention. Self-healing networks allow nodes to reconfigure their link associations and find alternative pathways around failed or powered-down nodes. How these capabilities are implemented is specific to the network management protocol and the network topology, and ultimately will determine the network s flexibility, scalability, cost and performance. Volume 3 Issue 3 March 2015 Page 6

2 2.2 Wireless Sensor Node A basic sensor node [2] comprises five main components (Figure 1.2): Controller A controller to process all the relevant data, capable of executing arbitrary code. Memory Some memory to store programs and intermediate data. Sensors and Actuators The actual interface to the physical world. Devices that can observe or control physical parameters of the environment. Communication Turning nodes into a network requires a device for sending and receiving information over a wireless channel. Power supply As usually no tethered power supply is available, some forms of batteries are necessary to provide energy. Sometimes, some form of recharging by obtaining energy from the environment is available as well (e.g. solar cells). 2.3 Types of Sensor Networks Current WSNs are deployed on land, underground, and underwater [3]. Depending on the environment, a sensor network faces different challenges and constraints. Terrestrial WSNs typically consist of hundreds to thousands of inexpensive wireless sensor nodes deployed in a given area, either in an ad hoc or in a pre-planned manner. Underground WSNs consist of a number of sensor nodes buried underground or in a cave or mine used to monitor underground conditions. Additional sink nodes are located above ground to relay information from the sensor nodes to the base station. An underground WSN is more expensive than a terrestrial WSN in terms of equipment, deployment, and maintenance. Underwater WSNs consist of a number of sensor nodes and vehicles deployed underwater. As opposite to terrestrial WSNs, underwater sensor nodes are more expensive and fewer sensor nodes are deployed. Autonomous underwater vehicles are used for exploration or gathering data from sensor nodes. Compared to a dense deployment of sensor nodes in a terrestrial WSN, a sparse deployment of sensor nodes is placed underwater. Multi-media WSNs have been proposed to enable monitoring and tracking of events in the form of multimedia such as video, audio, and imaging. Multi-media WSNs consist of a number of low cost sensor nodes equipped with cameras and microphones. These sensor nodes interconnect with each other over a wireless connection for data retrieval, process, correlation, and compression. Multi-media sensor nodes are deployed in a pre-planned manner into the environment to guarantee coverage. Mobile WSNs consist of a collection of sensor nodes that can move on their own and interact with the physical environment. Mobile nodes have the ability sense, compute, and communicate like static nodes. A key difference is mobile nodes have the ability to reposition and organize itself in the network. A mobile WSN can start off with some initial deployment and nodes can then spread out to gather information. Information gathered by a mobile node can be communicated to another mobile node when they are within range of each other. 2.4 WSN Applications [4] Habitat and Ecosystem Monitoring Environmental Application Civil Structural Health Monitoring Monitoring Groundwater Contamination Rapid Emergency Response Industrial Process Monitoring Military applications Automated Building Climate Control 2.5 Faults in WSN Faults are something which itself is unwanted and leads to unwanted results. As earlier said WSNs are failure prone due to any of the reasons like malicious attack, energy depletion, hardware failure, communication errors and so on. If the quality of network decreases then it proportionally affects the failure. Its effects vary from nothing to the total breakdown. Volume 3 Issue 3 March 2015 Page 7

3 2.5.1 Types of Faults The node status in WSNs can be divided as [5] : Node faults in WSNs can be divided into 2 types: Hard Faults: is when a sensor node cannot communicate with other nodes because of the failure of a certain module (e.g., energy depletion of node) Soft Faults: failed nodes can continue to work and communicate with other nodes, but the data sensed or transmitted is not correct Fault Management Process Fault management process is divided into three phases: Fault detection Fault diagnosis Fault recovery Necessity of Fault Detection Method Failed nodes may decrease the quality of service (QoS) of the entire WSN. It is important and necessary to study the fault detection methods for nodes in WSNs for the following reasons: It is not practical to manually examine the functionality of nodes. Due to harsh environment, failure in sensor node can occur more easily than in other systems. Nodes are usually battery-powered, so it is common for faults to occur due to battery depletion. Failed nodes can produce erroneous data which may result in collapse of entire network. WSNs are also deployed in high security regions such as monitoring of nuclear reactor. Fault detection in these cases has great importance Fault Detection: An Overview As mentioned, fault detection is the first step in fault management in which faults should be identified by network system. Article [6], divides the failure detection approaches into two categories: Centralized Approach: Centralized approach is a common solution to identify and localize the cause of failures in WSNs. Usually; a geographically or logically centralized sensor node takes responsibility for monitoring and tracing failed nodes in the network. The central node normally adopts an active detection model to retrieve states of the network performance and individual sensor nodes by periodically injecting requests into the network. It analyzes this information to identify and localize the failed or suspicious nodes. In addition, the central manager provides a centralized approach to prevent the potential failure by comparing the current or historical states of sensor nodes against the overall network information models. As a summary, the centralized approach is efficient and accurate to identify the network faults in certain ways. Distributed Approach: Distributed approach encourages the concept of local decision-making, which evenly distributes fault management into the network.the goal of it is to allow a node to make certain levels of decision before communicating with the central node. It believes the more decision a sensor can make, the less information needs to be delivered to the central node. In the other word, the control centre should not be informed unless there is really a fault occurred in the network. Node Self-Detection: A self detection model to monitor the malfunction of the physical components of a sensor node via both hardware and software interface has been proposed by number of researchers. Self-detection of node failure is somehow straightforward as the node just observes the binary outputs of its sensors by comparing with the pre-defined fault models. Neighbor Coordination: Failure detection via neighbor coordination is another example of fault management distribution. Nodes coordinate with their neighbors to detect and identify the network faults before consulting with the central node. In addition, a node can also query diagnostic information from its neighbors (in one-hop communication range). This allows the decentralized diagnostic framework to scale easily to much larger and denser sensor networks if required. Volume 3 Issue 3 March 2015 Page 8

4 2.6 METHODS FOR FAULT DETECTION It is desirable to identify, locate and elimination of faulty nodes from network. Otherwise it provides incorrect diagnosis. Here we provide a brief description of some fault detection algorithms with their references. [7] Concentrated on the head node performance in various clustering and minimizing the task of cluster head. A criterion is taken to identify faulty nodes using Poisson distribution that will observe the failure probability before the time of commencement. Wireless sensor network, in which nodes are conversed with multi-hop to transmit the data and pass them to sink node. The group of nodes can be molded as clusters; in which an explicit node is assumed to be cluster head. The head node pioneered based on its battery forte. They proposed a scheme of Poisson Failure Density Algorithm to conclude the results of evaluation with respect to set of parameters. To reduce the burden of cluster head, relay node is placed in between head and sink. The performance of relay node is better than the cluster node and cluster head. The probability of cluster head and other nodes before transmission begins is calculated and predict battery life time of nodes then replace with higher energy available to act as cluster head. Generally relay node can be used in various fields and it has more battery strength then normal sensor. IEEE standard for this device b/g and it consist of 16MB flash and 64MB SDRAM with 400 MHZ Linux system. relay nodes are in charge for data packet fusion from sensor nodes which in charge of sensing data from their clusters and transmitting them to destination node via wireless multi-hop paths. On the other hand, [8] presents a distributed, reference-free fault detection algorithm that is based on local pair-wise verification between sensors monitoring the same physical system. Specifically, a linear relationship is shown to exist between the outputs of a pair of sensors measuring the same system. Using this relationship, faulty sensors may be detected within subsystems of the global system. Compact and low-cost sensors used in wireless sensor networks are vulnerable to deterioration and failure. As the number and scale of sensor deployments grow, the failure of sensors becomes an increasingly paramount issue. An appealing feature of the proposed method is that the need for reference sensors and complete knowledge of the system input are not required. Due to the pair wise nature of the proposed algorithm, it can also be performed in a completely decentralized fashion. This ensures the method can be scaled to large sensor networks and lead to significant energy savings derived from reduced wireless communication compared to centralized approaches. A distributed algorithm for detecting and isolating faulty sensor nodes in wireless sensor networks is proposed in [9]. Nodes with malfunctioning sensors are allowed to act as a communication node for routing, but they are logically isolated from the network as far as fault detection is concerned. It employs local comparisons of sensed data between neighbors and dissemination of the test results to enhance the accuracy of diagnosis. Time redundancy is used to tolerate transient faults in sensing and communication. Peng Jiang[5] proposed an improved DFD scheme by defining new detection criteria. It is well known that the distributed fault detection (DFD) scheme checks out the failed nodes by exchanging data and mutually testing among neighbor nodes in this network., but the fault detection accuracy of a DFD scheme would decrease rapidly when the number of neighbor nodes to be diagnosed is small and the node s failure ratio is high. The improved DFD scheme can also be applied to wireless sensor networks where there are less neighbor nodes and the node failure ratio is higher. Transient faults in sensing and communication have been investigated in the paper [10]. A simple distributed algorithm has been proposed that tolerates transient faults in the fault detection process. Some other fault management schemes can also be found in this survey. They presented an online distributed diagnosis algorithm (CDFD) which is integrated with unequal cluster-based routing protocol. The diagnosis algorithm imposes a negligible extra cost in the WSN since diagnostic messages are sent as the output of the routine tasks of the WSN. Myeong-Hyeon Lee, Yoon-Hwa Choi said that a diagnosis is said to be a complete diagnosis if all the faulty nodes can be identified based on a given syndrome generated by the system. Similarly a diagnosis is said to be a correct diagnosis if, on the basis of a given syndrome, no fault-free nodes are identified as faulty. A complete and correct diagnosis is very difficult or sometimes might be impossible. Incomplete diagnosis in sensor networks could be acceptable if faulty sensor nodes determined to be fault-free can be isolated from the network of fault-free nodes and the number of them is manageably small. In addition, it is still safe to use a diagnosis algorithm that might be incorrect but can identify almost all of the faultfree nodes as long as a negligibly small number of fault-free nodes are excluded from the network. The reason for this is that sensor nodes are generally expected to be cheap and sufficient redundant nodes are typically deployed to achieve fault tolerance and sensing coverage.[9] Bhaskar Krishnamachari, Sitharama Iyengar [11] proposed solution, in the form of Volume 3 Issue 3 March 2015 Page 9

5 Bayesian fault recognition algorithms, exploits the notion that measurement errors due to faulty equipment are likely to be uncorrelated, while environmental conditions are spatially correlated. This is the first paper to propose a solution to the fault-event disambiguation problem in sensor networks. Theoretical analysis and simulation results show that percent of faults can be corrected using this algorithm, even when as many as 10 percent of the nodes are faulty. [12] gives contribution in the development of a generic localized fault detection algorithm for wireless sensor networks. Faults occurring to sensor nodes are common due to the sensor device itself and the harsh environment where the sensor nodes are deployed. They evaluate a localized fault detection algorithm to identify the faulty sensors. The implementation complexity of the algorithm is low and the probability of correct diagnosis is very high even in the existence of large fault sets. In distributed localized faulty sensor (DLFS) detection algorithm, each sensor identifies its own status to be either good or faulty and the claim is then supported or reverted by its neighbors as they also evaluate the node behavior. The proposed algorithm is analyzed using a probabilistic approach. At this time there may be issues related to scalability and overhead due to exchange of information between neighbors. REFERENCES [1] Chiara Buratti, Andrea Conti, Davide Dardari, and Roberto Verdone, An Overview on Wireless Sensor Networks Technology and Evolution , Sensors 2009 [2] Holger Karl and Andreas Willig, Protocols and Architectures for Wireless Sensor Networks, 2005 John Wiley & Sons, Ltd. ISBN: [3] Jennifer Yick, Biswanath Mukherjee, Dipak Ghosal, Wireless sensor network survey, (2008) ,Elseivier publications [4] Daniele Puccinelli and Martin Haenggi, Wireless Sensor Networks: Applications and Challenges of Ubiquitous Sensing, (2005), 19-31, IEEE Circuits And Systems Magazine [5] Peng Jiang, A New Method for Node Fault Detection in Wireless Sensor Networks, Sensors 2009, vol 9, [6] M. Yu, H.Mokhtar, M.Merabti, Fault Management in Wireless Sensor Networks, IEEE Wireless Communications (2007) [7] Sirajul Ameen.C, Mohammed Ashraf.A, Prabakaran.N, Fault Tolerance Using Cluster in Wireless Sensor Network,IJARCSSE (2014), Volume 4, Issue 4, [8] Chun Lo, Jerome P. Lynch, Mingyan Liu, Distributed Reference-Free Fault Detection Method for Autonomous Wireless Sensor Network, IEEE Sensors Journal, (2013) [9] Myeong-Hyeon Lee, Yoon-Hwa Choi, Fault detection of wireless sensor networks 31 (2008) ,Elseivier publications [10] Arunanshu Mahapatro, Pabitra Mohan Khilar Online Distributed Fault Diagnosis in Wireless Sensor Networks (2013) 71: , Springer Science Business Media NewYork 2012 [11] Bhaskar Krishnamachari, Sitharama Iyengar, Fault-Tolerant Event Region Detection in Wireless Sensor Networks 53(2004), IEEE Transactions for computers [12] Jinran Chen, Shubha Kher, and Arun Somani, Distributed Fault Detection of Wireless Sensor Networks, 2006 ACM, Volume 3 Issue 3 March 2015 Page 10