Dynamic energy efficient task offloading and resource allocation for NOMA-enabled IoT in smart buildings and environment

Abstract

Nowadays, the Internet of Things (IoT) devices are widely deployed and applied in smart buildings and environment, and play an important part in people’s daily life. With more and more IoT applications adopted, a number of computation-intensive tasks are generated. Due to the limitations of computation, storage resources , and battery capacities of IoT devices, some tasks need to be offloaded to edge servers for processing. Besides, the non-orthogonal multiple access (NOMA) techniques are also proposed to solve the problem brought by the limitation of wireless transmission resources. In this paper, we study the energy efficient task offloading and resource allocation problem for NOMA-enabled IoT in smart buildings and environment. We formulate the optimization problem with the goal of minimizing the energy consumption while meeting the delay constraints. However, solving this task offloading and resource allocation problem faces several challenges. Firstly, the statistical information of task generation process on IoT devices in smart buildings and the states of wireless channels are hard to predict precisely. Secondly, as the numbers of IoT devices and applications increase, the solution space of the optimization problem grows exponentially. To address these challenges, we take advantage of advanced stochastic optimization techniques, and transform the original problem into a deterministic problem. Then, we decompose the transformed problem into three subproblems and propose the dynamic energy efficient task offloading and resource allocation (NTORA) algorithm. Experiments include both parameter analysis and comparison experiments are conducted, and the results validate the effectiveness of our NTORA algorithm.