Energy-efficient network architecture for smart city development

Abstract

The growing number of objects connected to the Internet due to the evolution of smart cities development and the tremendous data in the future 6G networks with extensive processing has produced massive amounts of data that need to be processed. This will burden the conventional clouds to process the data besides reducing the Quality of Services due to high latency. Also, the increase in data volumes increased the energy consumed by transport networks and the cloud. Therefore, fog computing has been introduced to overcome the limitation of the cloud. This study proposed a new fog computing architecture for Internet-of-Things (IoT) applications based on wireless access networks using WiFi technology. A new mathematical model has been developed to optimize the number and locations of the fog servers at the access network to minimize the energy consumption of the networking and processing equipment at the access layer. This is beneficial for smart city development which contributes to the 2030 Agenda for Sustainable Development Goals (SDG) under Goal 7. The Mixed Integer Linear Programming (MILP) model using AMPL software with CPLEX solver is used to model the energy-efficient fog computing architecture considering twelve tourist locations in Melaka as the case study. The results show that the proposed optimized approach (OA) has 52.6% energy saving as the low number of fog servers and networking devices are utilized in the network compared to the non-optimized approach (NOA). Besides, the energy saving increased up to 81.1% when high processing server capacity is considered in the network. In addition to that, the results also show that increasing the traffic demands by each user does not give a significant energy increment when considering a fog server with high processing capacity in the network.