Article Preview
Top1. Introduction
The wireless sensor network may be thought of as a decentralized network system with several sensor nodes and central sink node(s) distributed over an area (Wang et al. 2017). A sensor node is made up of five major components: a sensing unit, digital data storage unit, energy supply unit, transceiver unit, and a limited computation unit. Navarro et al. report that particular types of sensors are attached to the nodes to sense the activities of the external environment. Depending on the application, many types of sensors are attached to the nodes. The sensing unit collects environmental data using various sensors and acts as the input device of the network. The sensor’s collected data is processed by the small processor, which then outputs the processed data (Sara et al. 2016). The data storage unit is used to store processed data in the local storage unit of the node. In general, the data storage capacity of WSN nodes is quite low (Ren et al. 2018). The transceiver unit sends and receives a limited quantity of processed data to nearby nodes or the sink node. The energy supply unit is responsible for supplying power to each unit of the WSN node. This self-contained battery power source is used to keep the node operational (Visconti et al. 2016). The sink node performs the function of the processing center. To build the network and establish a communication link, the sink node can communicate with each node. It sends and receives data to and from the dominating nodes. This type of network system is classified as a multi-hop communication system (Kumar et al. 2018).
In the modern era, Wireless Sensor Network (WSN) is also widely used in various fields like agriculture, meteorology, modern-day army, environmental monitoring, battlefield monitoring, body area network, intelligent household, etc. The WSN can be applied for many purposes: air pollution detection, fire detection, health monitoring, threat detection, etc. The WSN has a lot of applications in smart city projects also. The efficient implementation of WSN in a smart city can make the surveillance system which can reduce or stop unwanted events. As a result, a well-functioning efficient WSN is in high desire. But the design of the efficient WSN is not so easy. To configure a good WSN the hardware and software engineers should come up with experienced mathematicians or researchers having enormous knowledge of optimization and statistics. For many decades, optimization has been regarded as the most promising field of study, Bangyal et al. (2021). Optimization-based methods and statistical analytics are enormously used in the field of engineering. When a WSN system would be designed then the obvious question would be the sustainability (how long WSN can be in functioning status) of the system. Sustainability can be defined as the probability of persistency. One of the parameters of sustainability in the WSN system is energy. Energy is very limited in WSN. If the network's energy consumption can be minimized, the system's durability and sustainability can be improved, allowing the WSN to operate for longer periods. So, in this research work, the minimization of energy consumption (one of the essential measures) in a WSN has been chosen to prolong the sustainability of the network. To do energy minimization a multi-objective optimization approach has been considered. The node deployment strategy also has a significant impact on total energy usage in network communication. In this research, two types of deployment strategies have been used (Random deployment and U pattern deployment) to observe and compare the impact of energy consumption on the network.
The statistical analysis also has been carried out on the obtained data of the optimization algorithm. Now the very important question is that how a statistical tool can help in WSN design? Using this statistical analysis, the researchers can easily select the right objective function value for the energy minimization problem. The statistical approach will also enable the developer to make the right decision on hardware implementation. To determine if the findings of the proposed method were statistically significant, the authors, Seyedali Mirjalili et al (2014), utilized the Wilcoxon Rank-Sum statistical test. As a part of statistical analysis, the statistical tool, the Wilcoxon Rank-Sum test, has been used in this research.
The main aim of this research paper is to prolong the sustainability of the network by minimizing the network's overall energy consumptions. The research methodology can be briefly described in the following way: