Optimization of edge server group collaboration architecture strategy in IoT smart cities application

Abstract

With the development of big data and communication technologies, the Internet of Things (IoT) has permeated all aspects of smart cities. IoT smart city application scenarios are distributed with a large number of edge servers to accomplish large-scale data collection, transmission, analysis, and decision-making. However, in many emergency services, network communication faces data congestion and insufficient computational resources for nodes. To alleviate the situation that some nodes operate efficiently with insufficient cache and resource shortages, edge servers need to collaborate to handle tasks together and form an edge service community to realize fast message reception, response, and processing. Based on this, this study proposes an optimized edge server group collaboration architecture strategy in IoT smart cities application (ESGCA). It is based on the coalition to accomplish the optimal edge service community generation to collaborate on the messaging task. We design a multivariate discrete particle swarm optimization algorithm based on the discrete nearest past position update policy to improve the search utility. The algorithm can effectively solve the problem that current algorithms are prone to falling into local optimal solutions, long running times, and instability in the case of too many transmission tasks and edge nodes. Experimental results show that in the environment of insufficient node cache space and urgent transmission tasks, our ESGCA method can equalize the energy consumption of nodes, conserve computational resources, reduce the message transmission delay and the data loss rate.