Study of Data fusion in Wireless Sensor Network Kiran Maraiya#, Kamal Kant*, Nitin Gupta# η
Department of Computer Science and Engineering National Institute of Technology Hamirpur, Hamirpur (H.P.) INDIA 1
[email protected] 3
[email protected]
Ύ
Department of Computer Science and Engineering ASET, Amity University, Noida (U.P) ,India 2
[email protected]
Abstract— Wireless Sensor networks consist of a large number sensor nodes that are deployed in some geographical area. The purpose of the network is to sense the environment and report what happens in the area it is deployed in. With various potential application wireless sensor network has many challenges. Energy is the main constraint of the wireless sensor network due to limited battery power supplied to each sensor node, so it is necessary to incorporate energy awareness into every stage of the network design and operation. In order to maximize the lifetime of sensor networks, the system needs aggressive energy optimization techniques, ensuring that energy awareness is incorporated not only into individual sensor nodes but also into groups of cooperating nodes and into an entire sensor network. This paper gives a data fusion(aggregation) techniques to play very important role for enhance the accuracy and robustness of the information which obtain by entire network. With the explanation of clustering in WSN this paper gives the architecture overview for data fusion in sensor network. Keywords— WSN, Data Fusion, Energy, Cluster
I. INTRODUCTION A wireless sensor network is a type of network. It is basically a collection of small and tiny devices these small device called wireless sensor node or sensor node,which is monitor or record a physical environment situation such as temperature ,motion or pressure etc. Each node is a computer with close sensors that can development, replace sensing data, as well as communicate wirelessly among themselves to execute various responsibilities. Sensor networks are used to evaluate high temperature or pressure, or it could be used for object tracking or margin surveillance. It could be also deployed in factories in order to monitor poisonous or dangerous resources. It is also used to evaluate the weakness in building structures, or in vehicles and airplanes. Figure 1 show the architecture of a sensor node, this architecture consists 4 major parts. Energy supply, sensor and analog to digital converter (ADC), processor and memory, at last the transceiver .Basically transceiver are used for send and receive data simultaneously as shown in Figure 1 The Energy supply is to power the sensor node. The sensor circuitry can convert physical quantities into an electric signal. The ADC changes the analog signal generated by the sensor into a digital signal and sends it to the processor. The processor can execute easy operations on the received digital signal, and can store it into memory. at last, the transceiver sends and receives
data. Basically sensor nodes directly connect to to the each other, as well as sensor nodes communicate with sink node or base station connect to the internet . User gets a information from internet.But if we talk about configuration the sensor network .it provide a two type of configuration mode . figure 2 show the configuration method of the wireless sensor network.. In both configurations the nodes are scattered in a geological region, the region is divided into clusters with a cluster head in each cluster. Nodes in each cluster communicate with the cluster head The cluster head collects the data and ahead it to the user. In (a) nodes use chaining in organize to communicate with the cluster head while in (b) nodes directly communicate with the cluster head (gateway) in its cluster,. Using chaining reduces the energy used in conduction, but increases the energy used in processing since each node should receive and forward the message to and from other nodes. A number of sensor networks may have extra than one level of aggregation. Characteristically, sensor networks workings in one of two modes. Permanent operation, or query mode. In permanent operation mode, the nodes are continuously sensing the atmosphere and transfer the data (or the processed data) to adjacent or a central node. In query mode, the node is typically powered down waiting for a command from a central node, or neighbouring node. When the node receives the commands. It collects data from the sensor, processes it and sends it to the requesting node. Wireless sensor networks have many efficiency applications in military and civil, which may be classified into three classes: data collection, surveillance, and Target tracking.[2] Sensor unit Sensor
Computational unit Processor
ADC Sensor
Transceiver Transmitter /Receiver
Memory
Energy unit Figure 1 Sensor node architecture
Proc. of the International Conference on Advanced Computing and Communication Technologies (ACCT 2011) Copyright © 2011 RG Education Society ISBN: 978-981-08-7932-7 535
Proc. of the International Conference on Advanced Computing and Communication Technologies (ACCT 2011)
in network. The very important feature of wireless sensor network is the correlated data problem. In the wireless sensor network, the data collected by neighboring sensor nodes are often quite similar which make possible to the development of routing and fusion techniques. With the help of these techniques we can reduce redundancy of the data and enhance the accuracy or energy efficiency of wireless sensor network. III. CLUSTERING IN WSN:
Cluster head Sensor node Figure 2 Two configuration of sensor node
II.Wireless sensor network challenges: Wireless sensor network assure a wide variety of application and realize these application in real world .planning or designing a new algorithm or protocol address some challenges, these are follows: A. Physical resource constraint: Limited battery power of sensor node is very important constraint in a wireless sensor network. The sensor network is collection of sensor node .The lifetime of sensor node is recognizing by the power supply. Therefore lifetime of sensor node is also recognizing by the power supply. Therefore energy(power) consumption is important design issue, individual sensor node can store amount of data so limited computation power and memory size is constraint that affects on storing data that’s why designing algorithm and protocol should be light weighted and simple. B. Ad – hoc deployment: Most of the application needs the ad hoc deployment of sensor node in a region. Sensor node is randomly deployed over the region without and infrastrure which needs the system from connection between the nodes. C. Quality of service: Some application of wireless sensor network is depend on time critical. Which means the data should be delivered with certain period of time from the instead it is sensed, otherwise the data will be careless that’s why this could be a quality of service parameter for some application.
Sensor node are densely deployed in wireless sensor network that means physical environment would produce very similar data in close by sensor node and transmitting such type of data is more or less redundant. So all these facts encourage using some kind of grouping of sensor nodes such that group of sensor node can be combine or compress data together and transmit only compact data . this can reduce localized traffic in indidual group and also reduce global data .this grouping process of sensor nodes in a densely deployed large scale sensor node is known as clustering. The way of combing data and compress data belonging to a single cluster called data fusion (aggregation). Issues of clustering in wireless sensor network:1. How many sensor nodes should be taken in a single cluster. Selection procedure of cluster head in a individual cluster. 2. Heterogeneity in a network, it means user can put some power full nodes, in term of energy in the network which can behave like cluster head and simple node in a cluster work as a cluster member only. Many protocol and algorithm have been proposed which deal with each individual issue. IV. DATA AGGREGATION The aim of data aggregation(fusion) is that eliminates redundant data transmission and enhances the life time of energy in wireless sensor network. Data fusion is the process of one or several sensors then collect the detection result from other sensor .the collected data must be processed by sensor to reduce transmission burden before the are transmitted to the base station or sink. Redundant data
D. Scalability: Depending upon the application sensor node in wireless sensor network are deployed in a region. These sensor node could be hundreds, thousand or more, so the algorithm or protocol must scalable. E.Fault tolerance: In the hostile environment a sensor node may fail due to physical damage .if sensor nodes fails, the algorithm or protocol are working upon must accommodate these changes
536
Sensor node 1
Sensor node 2
Complement data Figure 3 sensor nodes have some redundant data
Proc. of the International Conference on Advanced Computing and Communication Technologies (ACCT 2011)
The simplest data fusion function is duplicate suppression if two source both send same data, Data fusion node will send only of these far ward. Data fusion is very necessary in wireless sensor network because sensor node have a capability of sense data after sensing it. they are transmit sense data to a base station or sink .it is basically direct transmission and it is expensive since base station may be located very far away and sensor nodes in a network needs more energy power to transmit data over long distance .so that better schemes is that fewer nodes transmit data to this far distance .these fewer nodes called cluster head of individual cluster in wireless sensor network. A. Advantage and Disadvantage of Data aggregation in wireless sensor network:Advantage: With the help of data aggregation process we can enhance the robustness and accuracy of information which is obtained by entire network, certain redundancy exists in the data collected from sensor nodes thus data fusion processing is needed to reduce the redundant information. Another advantage is those reduces the traffic load and conserve energy of the sensors. Disadvantage: The cluster head means data fusion nodes send fuse these data to the base station .this cluster head or fusion node may be attacked by malicious attacker. If a cluster head is compromised, then the base station (sink) cannot be ensure the correctness of the fusion data that has been send to it. Another drawback is existing systems are several copies of the fusion result may be sent to the base station (sink) by uncompromised nodes .It increase the power consumed at these nodes. B. Performance measure:There are very important performance measures of data fusion algorithm. These performances are highly dependent on the desired application. 1. Network lifetime: The network lifetime is define the number of data fusion rounds. Till the specified percentage of the total nodes dies and the percentage depend on the application .If we talk about some application,simultenously working of the all the sensor nodes is crucial hence the lifetime of the network is number of round until the first nodes which improves the energy efficiency of nodes and enhance the lifetime of whole network. 2. Latency: Latency is evaluate data of time delay experiences by system, means data send by sensor nodes and received by base station(sink).basically delay involved in data transmission, routing and data fusion. 3. Communication overhead: It evaluates the communication complexity of the network fusion algorithm. 4. Data accuracy: It is a evaluate of ratio of total number of reading received at the base station (sink) to the total number of generated. V. Data aggregation approaches 1. In-Network Aggregation:-
Wireless sensor network has distributed processing of sensor node data. Data aggregation is the technique. It describes the processing method that is applied on the data received by a sensor node as well as data is to be routed in the entire network. In which reduce energy consumption of the sensor nodes and also reduce the number of transmission or length of the data packet. Elena Fosolo et al in [5] describe “In network aggregation is the exclusive process of collecting and routing information through a multi hop network. Processing of data packet with the help of intermediate sensor nodes. The objective of this approach is increase the life time of the network and also reduces resource consumption. There are two type of approach for in network aggregation. With size reduction and without size reduction .In network aggregation with size reduction. It is the process in which combine and compressing the data received by a sensor node from its neighbors in order to reduce the length of data packet to be sended towards base station. Example, in some circumstance a node receives two data packets which have a correlated data. In this condition it is useless to send both data packets. then we apply a function like MAX, AVG, MIN and again send single data packet to base station. With help of this approach we reduce the number of bit transmitted in the network and also save a lot of energy .In network aggregation without size reduction is define to the process of data packets received by different neighbors in to a single data packet but without processing the value of data .this process also reduce energy consumption or increase life time of the network. Most of the data aggregation techniques fall under three techniques .such as tree based approachs, multipath approaches, and cluster based approaches. There also some hybrid approaches that combines any of the three techniques above. 2. Tree-Based Approach:The tree based approach is defining aggregation from constructing an aggregation tree. The form of tree is minimum spanning tree, sink node consider as a root and source node consider as a leaves. Information flowing of data start from leaves node up to root means sink(base station) . Disadvantage of this approach, as we know like wireless sensor network are not free from failure .in case of data packet loss at any level of tree, the data will be lost not only for single level but for whole related sub tree as well. This approach is suitable for designing optimal aggregation techniques’. Madden et al. in [6] data centric protocol know as Tiny aggregation(TAG) approach. The working of TAG is depend on two phases: distributed phase and collection phase. 3.Multi-path Approach:The drawback of tree based approach is the limited robustness of the system. To overcome this drawback , a new approach
537
Proc. of the International Conference on Advanced Computing and Communication Technologies (ACCT 2011)
was proposed by many researchers .in which sending partially aggregated data to single parent node in aggregation tree,a node could send data over multiple paths. In which each and every node can send data packets to its possibly multiple neighbors. Hance data packet flow from source node to the sink node along multiple path ,lot of intermediate node between source node to sink node so aggregation done in every intermediate node. using this approach we will make the system robust but some extra overhead.the example of this approach like ring topology, where network is divided in to concentric circle with defining level levels according to hop distance from sink.[3]propose a new strategy have both issues : energy efficiency and robustness. In which single path to connect each node to the base station it is energy saving but high risk of link failure. But on the other head multipath approach would require more nodes to participate with consequent waste of energy. Authors present a clever use of multi-path only when there is loss of packet which is implemented by smart caching of data at sensor nodes. Authors also argue that in many practical situation data may be gathered only from a particular region, so they use a different approach that relies on a spanning tree and provides alternative paths only when a malfunctioning is detected. Algorithm adopts a tree-based approach for forwarding packets through the network. In the ideal situation when no failures occur, this is certainly the best choice, as the minimum number of nodes is engaged in the transmission phase. In the presence of link or node failures, the algorithm will discover alternative paths, so as ensure the delivery of as many packets as possible within the time constraints. The problem with this approach is that it may cause the arising of hot spots and nodes along preferred paths will consume their energy resources quickly, possibly causing disconnection in the network. 4.Cluster-Based Approach:Cluster based approach is hierarchical approach. In clusterbased approach, whole network is divided in to several clusters. Each cluster has a cluster-head which is selected among cluster members. Cluster-heads do the role of aggregator which aggregate data received from cluster members locally and then transmit the result to base station(sink). K. Dasgupta et al. in [7] proposed a maximum lifetime data aggregation (MLDA) algorithm which finds data gathering schedule provided location of sensors node and base-station, data packet size, and energy of each sensor node. A data gathering schedule specifies how data packet are collected from sensors node and transmitted to base station for each round. A schedule can be thought of as a collection of aggregation trees. In [4], they proposed heuristic-greedy clustering-based MLDA based on MLDA algorithm. In this they partitioned the network in to cluster and referred each cluster as super-sensor. They then compute maximum lifetime schedule for the super-sensors and then use this schedule to construct aggregation trees for the sensors. W. Choi et al. in [1] present a two-phase clustering (TPC) scheme. Phase I of this scheme creates clusters with a cluster-head and each node
within that cluster form a direct connect with cluster-head. Phase I the cluster-head rotation is localized and is done based on the remaining energy level of the sensor nodes which minimize time variance of sensors and this lead to energy saving from unnecessary cluster-head rotation. In phase II, each node within the cluster searches for a neighbor closer than cluster-head which is called data relay point and setup up a data relay link. Now the sensor nodes within a cluster either use direct link or data relay link to send their data to cluster head which is an energy efficient scheme. The data relay point aggregates data at forwarding time to another data relay point or cluster-head. In case of high network density, TPC phase II will setup unnecessary data relay link between neighbors as closely deployed sensor will sense same data and this lead to a waste of energy. IV.BASIC ARCHITECTURE OF DATA FUSION IN WIRELESS SENSOR NETWORK 1. Centralized Architecture:Centralized architecture is very simplest architecture of wireless sensor network. In which we can apply data fusion process. means each sensor nodes sense a data and transmit to the one central node, called central processor fusion node .this central processor fuse the reports collected by all sensor nodes. In this architecture central node have a responsibility of whole network. The basic advantage of this architecture is it can be easily detected erroneous report of information which is taken by the entire wireless sensor network. The disadvantage is that inflexible to sensor changes and the workload is concerned at a single point.
Sensor node Cluster head Figure 4 Centralized architecture of wireless sensor network
2. Decentralized Architecture: The decentralized architecture of wireless sensor network, there is no single centralized node that makes decisions on behalf of all the sensor nodes. Data fusion occurs locally at each node on the basis of local observations and the information obtained from neighbouring nodes. in which all sensor nodes are connected to each other on the observation .The advantage of this architecture are scalable and tolerant to the addition or loss of sensing nodes or dynamic changes in the network.
538
Proc. of the International Conference on Advanced Computing and Communication Technologies (ACCT 2011)
Sensor node Figure 5 Dencralized architecture of wireless sensor network
5.Hierarchical Architecture: The hierarchical architecture of wireless sensor network is very important, In this architecture all sensor node are partitioned into hierarchical level. In the practical sensor network, the level 0 may contain many normal sensors organized in a topographical area, and to minimize the transmission power, the data from individual sensor nodes will be forwarded to all the distant fusion nodes by adopting a suitable routing algorithm and minimize the transmission power, the data of sensor can be forwarded to a fusion node through the nearly sensor nodes using routing algorithm like directed diffusion or simple flooding. The advantage of this architecture is that Workload is balanced among nodes in the wireless sensor network.
IIV.CONCLUSIONS The wireless sensor network consists of different sensor node that have very limited battery power, So main objective for maximize the lifetime of sensor network. This paper gives the overview of current challenges in the wireless sensor network, In WSN the energy is basically consumed by data transmission, and approximately 70% energy is consumed by data transmission so data transmission should be optimized in wireless sensor network, for maximizing the lifetime of network. Data transmission can be optimized with the help of effective protocol and effective ways of data fusion(aggregation). In this paper we discuss various techniques of data fusion (aggregation) and also give architectural overview and discuss the advantage and disadvantages of these techniques. With the help of these techniques we can reduce the transmission load and enhance the life time of the entire network. REFERENCES [1]
[2]
[3] BS
Level 2
Base station [4]
[5]
Level 1
[6] [7] Level 0
Cluster head Sensor node
Cluster
Figure 6 Hierarchical architecture of wireless sensor network
539
W. Choi, P. Shah, and S. K. Das, “A Framework for Energy-Saving Data Gathering Using Two-Phase Clustering in Wireless Sensor Networks”, in Proceedings of the International Conference on Mobile and Ubiquitous Systems: Networking and Services (MobiQuitous), Boston, 2004, pp. 203-212. H. Cam, S. Ozdemir, P. Nair, and D. Muthuavinashiappan, “ESPDA: Energy-Efficient and Secure Pattern-based Data Aggregation for Wireless Sensor Networks”, in Proceedings of IEEE Sensor- The Second IEEE Conference on Sensors, Toronto, Canada, Oct. 22-24, 2003, pp. 732-736.. L. Gatani, G. Lo Re, and M. Ortolani, “Robust and Efficient Data Gathering for Wireless Sensor Networks”, in Proceeding of the 39th Hawaii International Conference on System Sciences – 2006... K. Dasgupta, K. Kalpakis, and P. Namjoshi, “An Efficient Clusteringbased Heuristic for Data Gathering and Aggregation in Sensor Networks”, IEEE 2003.. E. Fasolo, M. Rossi, J. Widmer, and M. Zorzi, “In-Network Aggregation Techniques for Wireless Sensor Networks: A Survey”, IEEE Wireless communication 2007. S. Madden et al., “TAG: a Tiny Aggregation Service for Ad-hoc Sensor Networks,” OSDI 2002, Boston, MA, Dec. 2002. K. Dasgupta et al., “Maximum Lifetime Data Gathering and Aggregation in Wireless Sensor Networks”, In Proc. of IEEE Networks’02 Conference, 2002.
