Answers are grounded in this paper and cite their sources.
The application of Mobile Edge Computing (MEC) significantly enhances real-time data processing in high-concurrency environments. However, a crucial challenge in MEC is optimizing computation offloading and resource allocation under dynamic conditions. To address this, we propose mooDRLL, an online offloading algorithm that combines deep reinforcement learning with LSTM, leveraging LSTM’s ability to capture temporal dependencies in wireless channel conditions for more accurate decision-making. Furthermore, we employ the Newton method in the resource allocation phase to ensure rapid convergence and high precision under convex constraints. To reduce complexity while maintaining solution quality, we introduce an adaptive method that dynamically modifies the parameters of the mooDRLL algorithm in real time. Experimental results show that mooDRLL significantly outperforms traditional methods, achieving up to two orders of magnitude reduction in CPU execution latency.
No takes yet. Share an insight, caveat, or question.
Li et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: