On the Partial Offloading to Multiple Helpers-Based Task Offloading for IoT Networks

Ubiquitous computing is a key component of future wireless communications and Internet of Things (IoT) applications. By using distributed fog computing, data obtained from sensors can be timely processed and different tasks can be computed efficiently. As a result, IoT applications can make timely decisions and improve application reliability. In this paper, we consider a Partial Offloading to Multiple Helper (POMH) scenario where IoT device divide their tasks into subtasks that can be computed in a parallel manner. We propose a modified many-to-many matching-based task offloading algorithm to reduce the task latency in a POMH scenario. To overcome the externalities due to the dynamic preference profile of fog nodes, an intelligent proposal-based scheme is also introduced. The performance analysis of the proposed techniques shows that they provide quicker task computation as compared to different techniques available in the literature.