Task offloading in mobile fog computing by classification and regression tree

Fog computing (FC) as an extension of cloud computing provides a lot of smart devices at the network edge, which can store and process data near end users. Because FC reduces latency and power consumption, it is suitable for the Internet of Things (IoT) applications as healthcare, vehicles, and smart cities. In FC, the mobile devices (MDs) can offload their heavy tasks to fog devices (FDs). The selection of best FD for offloading has serious challenges in the time and energy. In this paper, we present a Module Placement method by Classification and regression tree Algorithm (MPCA). We select the best FDs for modules by MPCA. Initially, the power consumption of MDs are checked, if this value is greater than Wi-Fi's power consumption, then offloading will be done. The MPCA's decision parameters for selecting the best FD include authentication, confidentiality, integrity, availability, capacity, speed, and cost. To optimize MPCA, we analyze and apply the probability of network's resource utilization in the module offloading. This method is called by (MPMCP). To evaluate our proposed approach, we simulate MPCA and MPMCP algorithms and compare them with First Fit (FF) and local mobile processing methods in Cloud, FDs, and MDs. The results include the power consumption, response time and performance show that the proposed methods are superior to other compared methods.