A task-driven offloading algorithm based on Balanced Multi-Agent Deep Deterministic Policy Gradient (BMADDPG) that reduces average task processing latency by approximately 22.67% and decreases total system cost by at least 18.32% under high-load scenarios.
Abstract
With the development of the Internet of Things and multi-access edge computing, latency-sensitive tasks impose higher requirements on network computing capability and service quality. Existing task offloading methods usually focus on a single performance metric and often adopt time-slot-driven decision mechanisms, which may introduce additional waiting latency and increase system cost. To address these problems, this paper proposes a task-driven offloading algorithm based on Balanced Multi-Agent Deep Deterministic Policy Gradient (BMADDPG). First, a four-layer integrated air–space–ground multi-access edge computing network model is constructed, consisting of ground devices, unmanned aerial vehicles (UAVs), low-Earth-orbit (LEO) satellites, and cloud servers. Multi-level computing node collaboration is used to improve the system computing capability. Second, the task offloading process is modeled as a multi-agent Markov decision process, where ground user devices act as agents. A centralized Critic and distributed Actor structure is adopted for collaborative decision-making. The proposed algorithm uses DDPG to handle continuous action spaces and triggers offloading decisions immediately upon task arrival, thereby avoiding time-slot waiting overhead. Meanwhile, latency, energy consumption, and load balancing constraints are incorporated into the reward function to guide global resource allocation. Simulation results show that BMADDPG converges stably within about 800 training episodes. Compared with DDPG, DQN, PPO, and D3QN, it reduces average task processing latency by approximately 22.67% and decreases total system cost by at least 18.32% under high-load scenarios.
Mobile edge computing (MEC) supports computation-intensive and latency-sensitive Internet of Things (IoT) applications. However, collaborative task offloading in dynamic heterogeneous environments remains challenging due to coupled physical constraints, shared resource competition, and high-dimensional decision spaces. Existing multi-agent deep reinforcement learning (MADRL) approaches often rely on static penalties or centralized action truncation for constraint handling. These methods may lead to unstable training, conservative strategies, and limited collaboration. To address these limitations, this paper proposes a constraint-aware multi-agent edge collaborative offloading algorithm (CARE-CTDE). The offloading problem is formulated as a constrained Markov decision process and addressed under a centralized training and decentralized execution (CTDE) framework. Dynamic Lagrange multipliers replace fixed penalties to improve training stability and support smoother exploration near constraint boundaries. A multi-threshold-guided Lagrangian constraint regulation mechanism further coordinates heterogeneous constraints, including energy consumption, latency, and server capacity. In addition, a congestion-driven cost allocation method transforms global resource competition into dynamic cost signals, guiding agents toward more coordinated offloading decisions. The simulation results show that CARE-CTDE achieves better scheduling performance, resource utilization, and constraint satisfaction than baseline methods in dynamic heterogeneous MEC scenarios, demonstrating its effectiveness and robustness for constrained edge computing systems.
Yuxuan Yang, Hexing Wang, Yang Zhou· Mathematics· 0 citations
This paper addresses the joint task offloading and resource allocation problem in multi-user MEC systems and proposes a decentralized control framework based on Multi-Agent Reinforcement Learning (MARL), which achieves lower total system cost and faster convergence than the full-local, full-offload, and heuristic baselines.
Youssef Oukissou, Mohamed Amine Meddaoui, Ayoub Belaidi et al.· International journal of Com...· 0 citations
The rapid growth of Internet of Vehicles (IoV) applications has imposed strict requirements on low-latency and energy-efficient computing services. This letter investigates a multi-Uncrewed Aerial Vehicle (UAV)-assisted IoV system, where multiple Mobile Edge Computing (MEC)-enabled UAVs (MUs) collaboratively provide computing services for vehicular terminals (VTs). To improve service capability, we propose an energy-efficient task offloading and load balancing scheme that jointly considers vehicle mobility, task offloading and migration, and computing resource allocation to formulate an optimization problem. To solve this problem, a collective learning (CL)-enabled multi-agent reinforcement learning (CL-MARL) algorithm is proposed, where each agent learns optimal policies through centralized training and collective cooperative learning. Simulation results demonstrate that the proposed scheme outperforms benchmark strategies in terms of energy efficiency, task completion rate, and load balancing.
Yongbin Wang, Peng Lin, Yan Liu et al.· IEEE Wireless Communications...· 0 citations
: Mobile edge computing (MEC) is an effective paradigm for supporting latency-sensitive and computation-intensive intelligent applications. However, in dynamic mobile-edge network scenarios, mobile terminals experience time-varying wireless links due to mobility. Tasks may also arrive unpredictably, while multiple terminals compete for limited edge resources. As a result, MEC systems may suffer from service congestion and unbalanced resource utilization, which increases end-to-end latency and energy consumption. This paper investigates cooperative task offloading in dynamic MEC networks. The considered system comprises one macro base station and multiple small base stations equipped with edge-computing resources. In each time slot, each mobile terminal selects a service option, determines the task offloading ratio, and chooses its transmit power for task uploading. This sequential decision process is formulated as a multi-agent problem with continuous action spaces. Under the centralized training and decentralized execution (CTDE) framework, the problem is further modeled as a decentralized partially observable Markov decision process (Dec-POMDP). Standard multi-agent soft actor-critic (MASAC) is not fully suitable for this problem. Its original action model does not handle bounded continuous actions well. Its exploration strength may also be unsuitable at different training stages. Frequent policy updates can further make training unstable when critic estimates are inaccurate. To address these issues, this paper develops an adaptive Beta-policy and delayed-update multi-agent soft actor-critic method, abbreviated as ABDMASAC. This method uses a Beta policy to model bounded actions. It adjusts the entropy coefficient during training and delays policy updates to reduce training oscillations. Experimental results show that, under a unified training budget and a consistent evaluation protocol, the proposed method achieves a better overall trade-off than the selected MASAC-backbone and on-policy MARL baselines under the considered simulation settings in terms of overall reward, average end-to-end latency, and average energy consumption. In the large-scale scenario, compared with MASAC, it improves the overall reward by 17.8%, reduces the average end-to-end latency by 18.0%, and lowers the average energy consumption by 11.4%.
Zheng Yao, Jie Liu, Changjun Deng et al.· Computers, Materials & C...· 0 citations
Integrated Terrestrial–Non-Terrestrial Networks (ITNTN), which combine terrestrial base stations (BSs), High-Altitude Platform Stations (HAPS), and Low-Earth Orbit (LEO) satellites, are key enablers of 6G communication and edge computing (EC) services. However, energy-limited BSs, particularly HAPS and satellites, pose significant sustainability challenges under continuous operation. To address this issue, we propose an on-demand EC server activation framework integrated with intelligent task offloading across ITNTN. A joint optimization problem is formulated to maximize task offloading success while satisfying energy and quality-of-service requirements. To solve it, we propose an online Q-learning policy that adaptively manages task offloading and EC server activation without prior knowledge of traffic dynamics. Simulation results show that the proposed method achieves superior task offloading success and energy efficiency compared to online heuristic and offline metaheuristic baselines. These findings highlight the importance of energyaware On-Off EC control for sustainable ITNTN systems.
Insaf Rzig, W. Jaafar, Safwan Alfattani· International Conference on...· 0 citations
Existing studies on unmanned aerial vehicle (UAV)-based Mobile Edge Computing (MEC) primarily focus on independent tasks that can be decomposed and executed in parallel. However, many practical applications involve interdependent tasks that are naturally modeled as directed acyclic graphs (DAGs). Although recent work has begun to address DAG-based task offloading in edge computing systems, most approaches assume direct, routing-free communication between users and edge servers, thereby neglecting the impact of network topology. In this paper, we investigate the joint task offloading and resource allocation problem for DAG-structured applications in multi-UAV-assisted MEC networks with arbitrary topology, where tasks may be offloaded to UAVs over multiple hops. We formulate the problem as a mixed-integer nonlinear optimization problem that jointly minimizes task execution time and energy consumption. To efficiently solve this problem, we propose a topology-aware task offloading algorithm that decomposes the original problem into tractable subproblems by ranking subtasks to satisfy DAG precedence constraints and iteratively optimizing resource allocation and offloading decisions. Simulation results demonstrate the promising performance of the proposed algorithm.
Ke Ma, Junfei Xie, Shangping Ren et al.· International Conference on...· 0 citations