Resource Efficient Clustering and Next Hop Knowledge Based Routing in Multiple Heterogeneous Wireless Sensor Networks

1. Introduction

Tiny sized sensor nodes with advanced technologies capable of measuring temperature, humidity, vibrations etc., will be useful for forecasting natural disasters, human health and many more applications. Sensor nodes with limited communication range, limited energy and limited computation capacity are considered as unreliable infrastructure for wireless sensor networks. Sensors are deployed in an area beyond reach of humans like dense forest, underground mine, battlefield surveillance, etc. (Ngoko, Cerin, Gianessi & Jiang, 2015). In such situations battery charge and battery replacement are not possible. Therefore, power consumption is a critical issue in wireless sensor networks and many researchers are working upon it to improve the sensor networks lifetime (Hsu, Ken, Slagter, Shih-Chang & Yeh-Ching, 2014; Mao, Qi, Lin et al., 2016) Cluster technology is used for load balancing and to save the energy in sensor networks. In clustering technology as shown in Figure 1, sensor nodes are organized in the form of clusters and each cluster has a sink node known as cluster head (CH). Rest of the nodes associated with that CH are referred to as cluster members (CM). A CM senses the data and forwards the data to associated CH. CH aggregates the data, computes the data and using multi hop or single hop technique forwards compressed data to base station. the base station stores the data on a cloud and clients can access the data from cloud anywhere anytime. CH aggregates the data, discards redundant & unusual data and compresses the data to save the energy. CMs only communicate with related CHs thus avoiding exchanging of redundant information among them to save more energy. Base station has no constraints of power & computation and proposed protocol considered Raspberry Pi as a base station (Maity & Park, 2016). Clustering is dynamic in nature and is formed at random interval of time. In cluster technology when all CMs transmit their data through a single CH then cluster head lose their energy in early stage (Karthikeyan, Prassanna &Neelanarayanan, 2015; Lin, Yang, Zhu & Peng, 2014). A sensor node as a CH performing various extra works such as aggregation, compression and routing loses its energy faster than other nodes. To prolong the network lifetime high-power nodes, play the role as CHs because these have more residual power compared to others. Each CM can participate in an algorithm to be elected as a CH and once CH is elected other CMs start their process to join CH based on distance, energy and other important parameters. Once a cluster is formed with its selected CH then data routing takes place, where CH forwards the data packets to next hop CH, resulting in balanced energy consumption. Lifetime of the high-power node is very crucial for prolonged operation of the network. In proposed algorithm CBERERP, the authors addressed these issues and compared with various energy efficient routing algorithms. Proposed paper deals with two problems related to optimization of energy resource.

• •

  Energy efficient clustering with adjustment among high power nodes (HPN), transmission distance and number of intermediate to reach till base station BS.

• •

  Next hop knowledge based routing algorithm in heterogeneous wireless sensor networks.

The paper is organized as follows. Section 2 describes related work related to proposed work. Section 3 provides the network model used in our technique. Section 4 presents the energy resource efficient clustering & routing protocol (CBERERP) in detail. Section 5 analyzes results of the protocol. Finally, Section 6 summarizes the performance of the algorithm through simulations and results.