Improving IIoT performance through dynamic configuration of edge computing/MWSN routing protocol

Abstract

With the rapid growth of the industrial Internet of things (IIoT) and the advent of fifth generation (5G) technology have made efficient data communication and processing increasingly crucial. Several requirements concerning IIoT and 5G like connectivity, scalability, energy efficiency, interoperability, security, and privacy, are crucial for modern industrial settings with numerous connected devices generating large volumes of data. This study presents two novel techniques to address these requirements in mobile wireless sensor networks (MWSNs), a key component of IIoT. These techniques are designed to handle the dynamic nature of mobile sensor nodes (MSNs) while considering resource constraints. The development features novel protocols designed to enhance the network lifetime in MWSNs, thereby improving the quality of service (QoS) for the entire network through two significant contributions. Firstly, a novel dual tier cluster-based routing protocol, DTC-BR, organizes the network into virtual zones. Each zone includes multiple cluster members (CMs) that collect data and a single cluster head (CH) that aggregates it. DTC-BR was evaluated in MATLAB against metrics such as energy consumption, network lifetime, and scalability. The efficiency of DTC-BR is highlighted in comparative results, showing a network lifetime increase of 6%, 21%, 25%, and 37% over dynamic directional routing (DDR), mobilityaware centralized clustering algorithm (MCCA), low energy adaptive clustering hierarchymobile energy efficient and connected (LEACH-MEEC), and low energy adaptive clustering hierarchy-mobile (LEACH-Mobile or LEACH-M) protocols respectively, particularly exhibiting efficiency in larger networks with a large number of sensor nodes (SNs). Secondly, a dynamic resource allocation based on real-time elastic approach, DRAREA, has been developed. This technique integrates DTC-BR and involves MSNs offloading data to CHs, which then transmit it directly to nearby edge servers (ESs). The offloading to ESs enhances computing efficiency, prolongs battery life, and optimizes energy use. MATLAB simulations have demonstrated the superiority of DRA-REA in task offloading and resource utilization, significantly improving network QoS and mobile IIoT devices energy conservation. It outperformed benchmarks like genetic algorithm based multi-edge collaborative computation offloading (GECO) and resource-agnostic microservice offloading (RAISE) in terms of queue size by more than 55% and execution latency by more than 22%. These techniques offer viable solutions to IIoT challenges like response time, battery life, bandwidth, and privacy. They effectively utilize resources and balance workloads across ESs, addressing the critical demands of IIoT applications and enhancing the overall performance of MWSNs. As the integration between clustering-based routing protocol and incorporating edge computing (EC) in resource management is currently in an exploratory stage, there is a significant lack of research in this area. Therefore, the potential of the results holds significant promise for driving advancements in the field of IIoT. The findings of this research provide a solid foundation for further exploration and innovation in resource utilization and system performance optimization within the IIoT domain.