Structured network can be classified as follows:
· Flat network: In flat networks, all the sensor nodes have equal properties and perform equal role. Mainly, Data centric routing is used in this type of network. In this method, base station sends a query by means of flooding to all the sensor nodes. If any node has the data matches with the query transmit the response back to the BS. The multi hop path can be used to perform data transmission. So that latency can be increased. Flooding and gossiping SPIN, directed diffusion, rumor routing, gradient based routing are some of the examples of routing protocols performing in flat networks.
· Hierarchical Networks: Hierarchical networks overcome the drawbacks of flat networks. It uses clustering, node heterogeneity and reservation based scheduling. In hierarchical (layered) architecture, nodes have higher energy are used to process and send the information only while nodes have lower energy are used to perform the sensing only. Hierarchical routing is an efficient way to save energy within a cluster and by performing data aggregation and fusion to decrease the number of transmitted packets to the BS. In Hierarchical structure, there is mainly two-layer routing where one layer is used to select cluster-heads and the other layer is used for routing.
Ø Tree based structure: A tree kind structure is formed to organize sensor nodes. It mainly consists of one root node (BS), intermediate nodes and leaf nodes. Leaf nodes are used to sense the data and the intermediate nodes are used to perform aggregation and transmit to the root node. The main goal is nothing but to build energy efficient tree which performing data aggregation and data transmission. EADAT (Energy-Aware Data Aggregation Tree), E- SPAN (energy-aware spanning tree) and TBC (tree based clustering) are some of the examples for routing in tree based structure.
Fig. 2 Tree Based Structure
Ø Cluster based structure: In this, various nodes are grouped into small cluster area. The leader node of a cluster area is called as cluster head (CH). All nodes in a cluster transmit their sensed information to corresponding CH and these nodes known as member nodes (MNs). CH manages the group communication with the BS. LEACH (Low Energy Adaptive Clustering Hierarchy) is one of the well-known techniques for clustering mechanism5.
Fig.3 Cluster based structure
Ø Grid based structure: In this, Network area divide into small and equal sized grid. Sensor nodes are placed in grid such a way that high energy nodes work as master nodes and low energy nodes work as normal nodes in a grid. Normal nodes are used to perform sensing only and master nodes are used for data aggregation and data transmission. There is also a grid head node which is head of all master nodes and communicates directly to BS (base station). Commonly grid based structure also employ sleep-awake mechanism of sensor nodes for improving network lifetime.
Fig.4 Grid based structure
Ø Chain based structure: It is one of the hierarchical methods of aggregation which forms chain architecture.
In chain architecture, each sensor node can make a communication with its neighbors and each get turn to be the leader for transmitting data to the BS. It uses a token passing approach. After getting the token, each node transmits the data to the aggregator node. Finally it reaches the BS. The well-known protocol is nothing but PEGASIS (Power-Efficient Gathering in Sensor Information Systems).
Fig.5 Chain based structure
Routing protocol based on hierarchical structure: Some commonly used routing protocols in these networks are as follows:
· LEACH: The low energy adaptive clustering hierarchy (LEACH), is one of the first adaptive clustering-based protocol is introduced by W R Heinzelman, A Chandrakasan, and H Balakrishnan6 . In LEACH, Randomly deployed nodes organize themselves into local clusters and each cluster has a local base station (cluster head) and many normal or member nodes. Member nodes send the data to cluster head (CH) and CHs send data to sink node (BS). The cluster head selection is based on randomized rotation. LEACH can provide as much as a factor of 8 reductions in energy dissipation compared with traditional routing protocols and doubling the network lifetime.
· PEGASIS: The power-efficient gathering in Sensor information systems (PEGASIS), a chain based routing protocol and an enhancement over LEACH is studied by S. Lindsey et. al7. Rather than classifying nodes in clusters, the algorithm forms chains of the sensor nodes. Each node transmits to and receives from only one closest node of its neighbors and adjusts the power of their transmission. In this way, nodes formed a chain. A highly improved lifetime (about 100-300%) can be achieved over LEACH. A global knowledge of network is necessary for above work otherwise it cannot be complete and this is the drawback.
· Tree based clustering (TBC) for energy efficient wireless sensor network is proposed by K T Kim, C H Lyu, S Soo Moon and H Y Youn8 . In this routing protocol, a root node is selected before data transmission. The root node is basically the cluster-head. Therefore, clustering is done before tree structure constructing. The tree construction consists of two steps. The first step is the determination of the tree level of each member node in the cluster, and the second one is the formation of the tree based on that. The distance of the member nodes to the CH is decided the height of the tree or number of levels in a tree. It is effectively reduces and balances the energy consumption among the nodes, and thus significantly extends the network lifetime compared to the existing methods such as LEACH & PEGASIS.
· GAF (Geographic Adaptive Fidelity): This protocol aims at optimizing the performance of wireless sensor networks by identifying equivalent nodes with respect to forwarding packets. Two nodes are considered to be equivalent when they maintain the same set of neighbor nodes and so they can belong to the same communication routes. Source and destination in the application are excluded from this characterization. a virtual grid is constructed. This grid is formed by cells whose size allows to state that all the nodes in one cell can directly communicate with the nodes belonging to adjacent cells and vice versa. In this way, the nodes in a cell are equivalent. Nodes identify equivalent nodes by the periodic exchange of discovery messages with the nodes in their cells. With the information contained in these messages, the nodes negotiate which one is going to support the communications. The other nodes will stay powered off. With this procedure, the routing fidelity is kept, that is, there is uninterrupted connectivity between communicating nodes. However, the elected node periodically rotates for fair energy consumption. To do so, the nodes wake up periodically9.
· GSSC (Geography-Informed Sleep Scheduling and Chaining Based Routing) algorithm in wireless sensor network is proposed by Poonam Lohan and Rajni Chauhan10. In this routing algorithm, Geographical information was used to choose one active node from nodes having same sensing information and other nodes was in sleep mode. Chaining based routing scheme was used to route the sensed data from active nodes to the base station to further reduce the energy consumption. This scheme extends the network lifetime by mixing geography informed sleep scheduling of sensor nodes and chain based routing concept.
In all these protocols, the performance is better when more than two structures are used. Although, these structure are faced complexity.
Comparison of Four Hierarchical Structures: On the basis of various factor that affecting routing in WSN, the comparison is shown in the table.
Grid-based (without sleep-awake mech.)
Table: Comparison of different hierarchical structured networks
Conclusion: This paper concludes that first structure based WSN is better than structure less WSN. Second, the usage only one hierarchical structure for any particular WSN application is not enough for desired performance. If more than two structures are used, there are chances to meet desire output. So Routing protocols used more than two structures are more energy efficient and have increased network lifetime.