An innovative cluster-based power-aware protocol for Internet of Things sensors utilizing mobile sink and particle swarm optimization

27.Jun.2023

Abstract
Over the past decade, the Internet of Things (IoT) has become a necessary technology increasingly applied in many fields of research and development, such as smart cities and homes, health, industry, agriculture, security, and surveillance. In IoT systems, the use of sensors is considered an important common manner in which all devices communicate with a wireless sensor network to form an information system that comprises a massive number of sensor nodes performing accurately to create a smart decision-making method. However, these sensor nodes might be employed in severe environments, where replacing or recharging their batteries is considered an impossible mission. Simultaneously, the limitation of energy resources in sensor nodes presents a challenging issue that reduces the lifespan of individual nodes and the overall network system as a result of energy depletion. These obstacles necessitate energy-efficient routing protocols. According to the literature review, various routing protocols have been introduced, especially those that use clustering techniques. However, they have many drawbacks due to the way of selecting the cluster head (CH), which results in consuming energy dramatically, and consequently shortening the network lifetime. Additionally, instead of using the static sink, which was inefficient in collecting data, many researchers studied the behavior of the mobile sink (MS), which also has many downsides that negatively impact network performance. This paper presents a novel energy-conscious protocol for clusters that incorporates an adaptive movement for mobile stations and utilizes particle swarm optimization (PSO). The circular network area is divided into clusters, each of which has an elected CH based on the PSO technique. The MS aims to distribute energy among nodes to prevent hotspot issues. To achieve better coverage, it moves in a circular pattern with a constant angular velocity, starting from the center of the network area and moving forward and backward along the radius of the network. This research conducts intensive simulations, which run on MATLAB R2018, to assess the performance of our proposed protocol and compare its results with those of pertinent works. The results obtained are encouraging and demonstrate that the protocol we proposed surpasses its counterparts by significantly extending the lifespan of the network.​