Latency-Aware Resource Allocation in Green Fog Networks for Industrial IoT Applications

Abstract

Green fog computing/networking is considered a promising way to get real-time communication for latency-sensitive Internet-of-Things (IoT) applications. Fog networks consist of fog/cloudlet nodes near the vicinity of users, served directly from the remote cloud server. The addition of cloudlet servers within IoT devices access scope gives rise to user interaction and power allocation problems. Within this paper, we have formulated a problem of joint resource and power allocation for fog networks enabled by the cache. The problem formulated is classified under the problem of NP-hard mixed-integer non-linear programming (MINLP). The goal of the proposed problem is to reduce the network's end-to-end energy in terms of overall network latency, under power, cache capacity, and Quality of Service (QoS) limitations. We are proposing the outer approximation algorithm (OAA) to achieve optimal performance of ε. The efficiency of the proposed algorithm is seen in terms of network key performance indicators (KPIs), such as throughput, number of associated devices and minimum criteria for data rates and latency.