Energy-efficient smart gateway framework with QoS-aware resource allocation in IoT ecosystem

Abstract

Energy-efficient smart gateways are the heart of an IoT ecosystem, as they are essential for efficiently handling incoming data and resources. Due to the exponential increase in IoT devices and applications, a vast amount of data is collected from the surrounding environment. This data needs to be processed and analyzed for further executions. Smart gateways play a crucial role in catering to user needs while handling vast amounts of data, and consuming less energy. Therefore, a smart gateway framework is proposed in this work to make dynamic decisions based on the user requirements while providing the best quality of service with efficient energy consumption. The proposed gateway framework is capable of handling urgent tasks that are latency-sensitive and also optimally allocating resources based on their computational needs in a fog-cloud environment. The incoming user-generated tasks are efficiently scheduled using the proposed Dynamic Priority-Multilevel Feed Back Queue Algorithm (DP-MFBQ). Whereas, the Machine Learning-based Resource Allocation (MLRA) algorithm was proposed to make dynamic decisions based on past learnings of the smart gateway. The proposed framework is simulated using the Yet Another Fog Simulator (YAFS) simulation toolkit, and it outperformed the existing work when evaluated on the Quality of Service parameters, including latency, wait time, energy consumption, and throughput.