Reinforcement learning for intelligent online computation offloading in wireless powered edge networks Ehzaz Mustafa 1 • Junaid Shuja 1 • Kashif Bilal 1 • Saad Mustafa 1 • Tahir Maqsood 1 • Faisal Rehman 1 • Atta ur Rehman Khan 2 Received: 31 January 2022 / Revised: 18 July 2022 / Accepted: 28 July 2022 Ó The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022 Abstract The method of charging mobile devices with wireless power transfer (WPT) from the base station (BS) integrated with mobile edge computing (MEC) increases the potential of MEC. The increasing demand for intelligent computation offloading requires effective decisions among local or remote computation specifically in wireless fading channels of the dynamic environment. Our main aim is to generate an effective offloading decision between local and remote computation in a real-time environment for each wireless channel while preserving optimal computation rate. In this article, we consider a wireless powered MEC system that governs a binary offloading decision to execute the task locally at the edge devices or the remote server. We propose a reinforcement learning based intelligent online offloading (RLIO) framework that adopts an optimal offloading action based on reinforcement methods. This framework acquires a worthy decision among local or remote computation for the time varying wireless channel conditions in dense networks. Numerical results show that the proposed framework can achieve optimal performance while preserving the computation time compared with existing optimization methods. Second, the average execution cost of RLIO is less than 0.4 ms per channel, which enables real-time and optimal offloading in dynamic and large-scale networks. Keywords Mobile edge computing  Wireless power transfer  Computation offloading  Reinforcement learning 1 Introduction Smart sensors and wireless devices have limited energy and require periodic charging, which is not possible in every scenario. Computational offloading and wireless power transfer (WPT) accommodate wireless devices to preserve operational energy levels. Computation offloading to the cloud servers minimizes the energy consumption of wireless devices [1, 2]. Computationally intensive appli- cations are offloaded to cloud servers to strengthen the capabilities of wireless devices and to save their battery power [3]. However, offloading computational tasks to cloud servers leads to high latency constraints [4–6]. To amend the latency limitations, a mobile edge computing (MEC) server is placed near the end wireless devices, which minimizes the transmission latency [7–9]. Compu- tational offloading in MEC has become an effective tech- nique to reduce the latency restraints that are linked with cloud computing [10–13]. Still, the energy and battery consumption of wireless devices limit the performance of MEC. To rectify the issue of battery drainage and energy consumption, WPT has been proposed as an effective approach. WPT can encourage wireless devices to offload computation tasks to nearby edge servers. Several researchers proposed joint WPT–MEC computational offloading to edge servers in a wireless environment [14, 15]. Figure 1 illustrates the mechanism of the WPT– MEC system in which an AP integrated with the MEC server transfers wireless power to wireless devices. These devices offload their compute-intensive tasks to edge ser- vers or compute locally with harvested energy. Multiple studies optimized the energy consumption of wireless devices with the help of MEC computational & Junaid Shuja junaidshuja@cuiatd.edu.pk 1 Department of Computer Science, COMSATS University Islamabad, Abbottabad Campus, Abbottabad, Pakistan 2 College of Engineering and Information Technology, Ajman University, Ajman, United Arab Emirates 123 Cluster Computing https://doi.org/10.1007/s10586-022-03700-5 Content courtesy of Springer Nature, terms of use apply. Rights reserved.