Contents I. Introduction
WSNs with hundreds or thousands of micro-sensor nodes can be deployed to support for wide range of applications in many different situations, such as battlefield surveillance, environmental monitoring, intelligent home and etc. The sensor nodes are equipped with small size, low cost, limited bandwidth, processor abilities and resources; particularly, small capacity battery of nodes cannot be recharged or replaced during operation time [1], [2]. Therefore, how to use energy efficiently is very important for designing routing protocols for WSN to maximize network lifetime. Cluster-based routing protocols [1], [3], [4] are widely known as a good technique for maintaining energy efficiency, which support to both homogeneous and heterogeneous network scheme, such as LEACH (Low Energy Adaptive Clustering Hierarchy) [2]–[4], PEGASIS [5], IEEPB [6] and so on. In LEACH protocol, nodes are organized into several clusters. Each cluster elects a leader node called CH, which is responsible for fusing many sensed data packets from its cluster member node(s) with its own data packet into a single packet and forwarding the fused packet to the BS; other nodes (cluster-members) will send sensed data to the respective CH by a single-hop mode, periodically. The energy of CH is rapidly exhausted because it has to further communicate and process more work than other nodes in cluster. Consequently, the role of CH must be passed to another node randomly after a certain time to balance energy consumption between nodes in the WSN. An improvement of LEACH algorithm was proposed by Kaur, et al called EE-TLDC (Energy Efficient Two Level Distributed Clustering) [7], in which the criteria for the selection of CH is based on probability and residual energy of candidate nodes in current time. Moreover, EE-TLDC reduces number of CHs, which transmit directly data packet to BS, by choosing in CHs list few SCHs. SCHs are responsible for forwarding data to BS, other CHs will transmit to nearest SCH instead of BS to save energy. However, the drawback of LEACH and EE-TLDC is that the single-hop communication between nodes and CH or BS, that is far, so, it will die quickly, although the algorithm decreases the complexity. Stephanie Lindsey et al proposed PEGASIS, which is a basic chain-based routing protocol [5], where sensor nodes only connect and communicate with the nearest neighbor into a chain. In order to transmit the fused data to BS, PEGASIS chooses a node to become CH in each round, which has random location in the chain. The simulation results show that performance of PEGASIS and IEEPB are better than LEACH [8], [9]; however, there are still some limitations in this protocol. Firstly, the CH is selected at random location in chain, (no considering the residual energy and distance to the BS). Secondly, some “long links” still exists due to simple formation chain algorithm. In addition, high delay or a bottleneck at the CH can occur in data transmission phase since the CH is a single node in long chain in PEGASIS. Up until now, there have been many chain-based routing protocols in homogenous network; they are improved based on PEGASIS such as IEEPB [10], EECB (Energy-Efficient Chain-Based) [9], and so on. However, none of the above improvements consider the time length of each round and how to balance number of nodes in each cluster. Moreover, most of them only deploy in homogeneous, not in a heterogeneous network, where two or more different types of sensor nodes are used those have different battery capacities and functions and it is similar to real situations more [7], [11]. With above analysis, in this paper, we propose Sector-Chain Based Clustering Routing Protocol, namely SCBC base on PEGASIS, which can achieve advantages of both EE-TLDC and IEEPB by dividing the network into logical sectors, which balance the number of nodes and sectors. In SCBC, the Greedy algorithm is used to form the chain alike IEEPB, but SCBC can avoid “long link” in chain by comparing the distance between nodes three time to find out node, which has the shortest link, to join chain. In addition, SCBC chooses CH, SCH in each round by considering remaining energy of candidate nodes and distance between them and BS to decide which node will become the CH or SCH. So, SCBC can enhance energy efficiency by calculating the time length in steady data transmission phase for each round. Our simulation results show that the network lifetime of SCBC can be extended to about 70% and 20% in comparison with PEGASIS and IEEPB, respectively. The rest of this paper is organized as follows. Section II presents the framework and Section III describes the detail of SCBC. In Section IV, evaluation and analysis of simulation results are presented. Finally, Section V presents our conclusion.
